Complaint Counsel s Post Trial Brief - In the Matter of Realcomp II Ltd.

UNITED STATES OF AMERICA BEFORE THE FEDERAL TRAE COMMISSION DOCKET NO. 9320 PUBLIC VERSION IN THE MATTER OF II REALCOMP LTD. COMPLAINT COUNSEL'S POST- TRIAL BRIEF Jeffrey Schmidt Director Sean Gates Kenneth L. Glazer Deputy Director Geoffrey Oliver Peggy Bayer Femenella Joel Christie Linda Holleran Chrstopher Renner Counsel Supporting the Complaint Assistant Director Patrck Roach Deputy Assistant Director Bureau of Competition Federal Trade Commission Washington , DC 20580 Dated: August 3 , 2007 TABLE OF CONTENTS I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 II. STATEMENT OF FACTS ............................................... 4 A. Real Estate Brokers and the Brokered Real Estate Transaction .............. 4 B. The Local MLS ................................................... 5 C. Realcomp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1. Realcomp's Governance and Membership ........................6 2. The Realcomp MLS Member Services ........................... 6 D. Real Estate Brokerage Service Models and Listing Contrcts. . . . . . . . . . . . . . . . 8 1. Traditional Full Service Brokers and Exclusive Right to Sell Listings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2. Limited Service Brokers and Exclusive Agency Listings ............ 10 3. Competition Among Full Service and Limited Service Brokers . . . . . . . 11 E. Realcomp's MLS Member Serices are Significant to Brokers' Ability to Compete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 1. "Exposure" of Listings is Important to Consumers and Critical to a Broker's Ability to Compete .................................. 13 2. Exposure Through the Realcomp MLS Database is Important For Brokers to Compete Effectively ...................................... 13 a. The Importance of the Local MLS . . . . . . . . . . . . . . . . . . . . . . . . 14 b. Access to the Realcomp MLS Is Important For Brokers in Southeastern Michigan To Compete Effectively. . . . . . . . . . . . . 16 3. Exposure Through Realcomp's Feed to the Approved Websites is Important For Brokers to Compete Effectively . . . . . . . . . . . . . . . . . . . . 17 a. Internet Marketing is Necessary to Reach Buyers Searching for Homes on the Internet ................................. 18 b. Internet Marketing is an Effective Means to Generate Leads .............................................. 19 c. Buyers Search for Homes on Four Key Categories of Websites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 d. The Competitive Signficance of Realcomp's Feed of Listing Information to the Approved Websites ....................22 F. Realcomp Adopted Policies That Limit the Exposure of Exclusive Agency Listings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 G. The Impact of Realcomp's Policies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26 Exclusive Agency Listings 1. The Realcomp Policies Limt the Exposure of to Buyers and Cooperating Brokers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 a. The Realcomp Website Policy Severely Limits the Exposure of Exclusive Agency Listings to Buyers .....................27 b. The Realcomp Search Function Policy Limited Exposure of Exclusive Agency Listing to Cooperating Brokers ........... 30 2. Realcomp's Policies Affect Discount Brokers' Ability to Compete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 3. Realcomp's Policies Reduced the Use of Discount Brokerage Services on a Market-Wide Basis ........................................ 34 H. Realcomp's Agreement to Change Its Search Function Policy but Refusal to Change its Website Policy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38 III. LEGAL ANALYSIS ................................................... 39 A. The Antitrst Laws Seek to Protect Competition to Ensure Lower Prices, Higher Quality, and Greater Choices for Consumers ...........................39 B. The Elements of Complaint Counsel's Claims and the Burden of Proof ......40 11 C. Realcomp Unreasonably Restrains Competition in the Real Estate Brokerage Services Market and Has Violated Section 5 of the FTC Act. . . . . . . . . . . . . . .45 1. Realcomp' s Policies are the Result of an Agreement Among Competing Real Estate Brokers ......................................... 45 2. The Nature of Realcomp's Policies is to Restrict Competition. . . . . . . .47 a. Realcomp's Policies Har Competition and Consumers By Restricting Competition from Discount Brokers . . . . . . . . . . . . . 47 b. Realcomp's Policies Harm Competition and Consumers By Eliminating a Product Desired By Consumers . . . . . . . . . . . . . . . 51 3. Realcomp Has the Power to Restrain Competition Among Real Estate Brokers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 a. "Network Effects" and Multiple Listing Services. . . . . . . . . . . .55 b. Market Definition. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 c. Realcomp's Market Shares Show that it Has Market Power ............................................. 56 d. Realcomp' s Market Power in the Multiple Listing Services Market Gives Realcomp the Ability to Restrain Competition in the Real Estate Brokerage Services Market . . . . . . . . . . . . . . . . . . . . . . . . 60 e. Other Evidence Confirms That Realcomp has Market Power and the Ability to Restrict Competition Among Brokers . . . . . . . . . . 60 4. Evidence of Actual Anticompetitive Effects Confirms that Realcomp's Policies Harm Competition and Consumers ...................... 63 a. Realcomp's Policies Substantially Reduced Limited Service Brokerage Activity ................................... 63 b. Realcomp's Policies Have Forced Consumers to Purchase Unwanted Services and to Pay More for Brokerage Services ............................................ 65 ii c. Realcomp's Policies Have Likely Maintained Higher Brokerage Fees and Reduced Output of Brokerage Services .............................................65 d. The Evidence Establishes that Realcomp's Policies Have Had Substantial Actual Antiticompetitive Effects. . . . . . . . . . . . . . . . 67 5. Realcomp's Policies Lack Any Pro competitive Justification .........68 iv. REMEDY. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 A. Realcomp' s Agreement to Change its Search Function Policy . . . . . . . . . . . . . . 73 B. To Restore Competition, the Cour Must Enjoin Realcomp from Enforcing Any Rules that Discriminate Against Brokers Using Exclusive Agency Listings ... 74 V. CONCLUSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 iv TABLE OF AUTHORITIES Federal Cases Alled Tube & Conduit Corp. v. Indian Head, Inc., 486 U.S. 492 (1988) ...................... 44 Alvord-Polk, Inc. v. F. Schumacher & Co., 37 F.3d 996 (3rd Cir. 1994) ........................ 44 In re American Industrial Real Estate Association, 116 F.T.C. 704 (1993) ..................... 1 Associated Press v. United States, 326 U.S. 1 (1945) .................................................... 45 In re Bellingham-Whatcom County Multiple Listing Bureau, 113 F.T.C. 724 (1990) ..... 1 In re Brunswick Corp., 94 F.T.C. 1174 (1979) ............................................................... 43 California Dental Associate v. FTC, 526 U.S. 756 (1999) ...................................... passim Cantor v. Multiple Listing Service, Inc, 568 F. Supp. 424 (S.D.N.Y. 1983) .............. 1,50 Chicago Professional Sports Ltd. Partnership v. National Basketball Association, 961 F.2d 667 (7th Cir. 1992) ........................................................... 68,69 Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752 (1984) .......................... 44 FTC v. Indiana Federation of Dentists, 476 U.S. 447 (1986) ................................. passim FTC v. Ruberoid Co., 343 U.S. 470 (1952) .................................................................... 72 FTC v. Staples, Inc., 970 F. Supp. 1066 (D.D.C. 1997) ................................................. 54 Fashion Originators' Guild, Inc. v. FTC, 312 U.S. 457 (1941) ...................................... 40 Flegel v. Christian Hospital, Northeast-Northwest, 4 F.3d 682 (8th Cir. 1993) ............. 61 Ford Motor Co. v. United States, 405 U.S. 562 (1972) .................................................. 70 General Leaseways, Inc. v. National Truck Leasing Association, 744 F.2d 588 (1984) ........ ........ ............................... ............ ........ ............................. ................. 53 Glen Holly Entm't Inc. v. Tektronix Inc., 352 F.3d 367 (2003) ................................ 39,65 Gordon v. Lewiston Hospital, 423 F.3d 184 (3rd Cir. 2005) ........................................... 60 Hahn v. Oregon Physicians' Service, 868 F.2d 1022 (9th Cir. 1988) .............................. 40 Image Tech. Services v. Eastman Kodak Co., 125 F.3d 1195 (9th Cir. 1997) ........... 66,67 Kreuzerv. American Acad. of Periodontology, 735 F.2d 1479 (D.C. Cir. 1984) ........... 48 Law v. NCAA, 134 F.3d 1010 (10th Cir. 1998) .......................................................... 39,43 Massachusetts Board of Registration in Optometry, 110 F.T.C. 549 (1988) ................. 41 In re Metropolitan MLS, Inc., No. C-3286, 1990 WL 10012611 (1990) ......................... 1 In re Multiple Listing Service of the Greater Michigan City Area, Inc., 106 F. T .C. 95 (1985) ....... ...;............................ ............ ....... ........................... ..................... 1 NCAA v. Board of Regents of the University Okla., 468 U.S. 85 (1984) ............ 39,42,53 In re North Texas Specialty Physicians, Dkt No. 9321, 2005 FTC LEXIS 173, at 36 (F.T.C. Nov. 29, 2005) ............................................................................. 40 Northern Pacifc Railway Co. v. Us., 356 U.S. 1 (1958) ............................................... 38 Northwest Wholesale Stationers, Inc. v. Pacifc Stationery & Printing Co., 472 U.S. 284 (1985) ........................................................................................... passim In re Orange County Board of Realtors, Inc., 106 F.T.C. 88 (1985) .............................. 1 In re Polygram Holding, Inc., 2003 FTC LEXIS 120 (JuI24, 2003) .............................. 43 Polygram Holding, Inc. v. FTC, 416 F.3d 29 (D.C. Cir. 2005) ................................ 41,42 In re Port Wash. Real Estate Board, Inc., 120 F.T.C. 882 (1995) .................................... 1 In re Puget Sound Multiple Listing Service, 113 F.T.C. 733 (1990) ................................ 1 Rebel Oil Co. v. Atlantic Richfeld Co., 51 F.3d 1421 (9th Cir. 1995) ............................ 54 Reifert v. South Central Wis. MLS Corp., 450 F.3d 312 (7th Cir. 2006) ......................... 58 Standard Oil Co. v. United States, 283 U.S. 163 (1931) ................................................ 40 Tampa Electric Co. v. Nashvile Coal Co., 365 U.S. 320 (1961) ................................... 54 Thompson v. Metropolitan Multi-List, Inc., 934 F.2d 1566 (1991) ...... ................................. ................... ...................... ......... passim Toys "R" Us, Inc., 126 F.T.C. 415, affd, 221 F.3d 928 (7th Cir. 2000) .............. 64,65,68 United Air Lines, Inc. v. Civil Aeronautics Board, 766 F.2d 1107 (7th Cir. 1985) ......... 49 11 In re United Real Estate Brokers of Rockland, Ltd., 116 F.T.C. 972 (1993) .................... I United States v. Brown University, 5 F.3d 658 (3rd Cir. 1993) ................................. 42, 60 United States v. Dentsply International, Inc., 399 F.3d 181 (3rd Cir. 2005) ................... 66 United States v. E.I. du Pont de Nemours & Co., 351 U.S. 377 (1956) ......................... 53 United States v. E. 1. du Pont de Nemours & Co., 366 U.S. 316 (1961) ........................ 72 United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001) .............................. 54,55 United States v. National Association of Real Estate Boards, 339 U.S. 485 (1950) ........ 1 United States v. Parke, Davis & Co., 365 U.S. 125 (1961) ............................................ 37 United States v. Realty Multi-List, Inc., 629 F.2d 1351 (5th Cir. 1980) ............................................................................. passim United States v. Uniroyal, Inc., 300 F. Supp. 84 (S.D.N.Y. 1969) ................................. 37 United States v. United Shoe Machine Corp., 391 U.S. 244 (1968) ............................... 70 United States v. United States Gypsum Co., 340 U.S. 76 (1950) ................................... 70 United States v. VISA US.A., Inc., 344 F.3d 229 (2nd Cir. 2003) ................................... 51 Welchlin v. Tenet Healthcare Corp., 366 F. Supp. 2d338 (D.S.C. 2005) ...................... 39 State Cases Austin Bd. of Realtors v. E-Realty, Inc., 2000 WL 34239114 (W.D. Tex. Mar. 30, 2000) .. ............... ................... ........... ........ .................... .................... 1 Grilo v. Board of Realtors, ,219 A.2d 635 (1966) .......................................................... 1 Marin County Board of Realtors v. Palsson, 16 Ca1.d 920 (1976) ................... 46,47,50 Oates v. Eastern Bergen County Multiple Listing Service, Inc., 273 A.2d 795 (1971) ....................................................................................... 1,46,48 Miscellaneous Federal Trade Commission and U.S. Department of Justice, Competition in the Real Estate Brokerage Industry at 12 (April 2007) ........................................ 13 ii I. INTRODUCTION In the early 2000s, full-service real estate brokers in Southeastern Michigan faced a dilemma. Brokers offering discounted, limited services were making inroads into the industr. These limited service brokers represent a new and importnt form of competition that puts price pressure on traditional, full-service commissions. They offer consumers substantial discounts by using a particular form of real estate contract, an "Exclusive Agency" agreement, instead of the traditional "Exclusive Right to Sell" agreement used by full-servce brokers. Although the fullservice brokers controlled the critical means for brokers to compete - Realcomp, the local multiple listing service - they were advised by more than one legal counsel that they could not simply ban the use of the limited service brokers' key weapon, the Exclusive Agency agreement. Unable to impose an outrght ban (which the Commission has long condemned as anticompetitive), i the full-service brokers found another way to use the MLS to squelch competition.2 Through Realcomp, they imposed discriminatory policies to disadvantage brokers i See, e.g., In re Port Wash. Real Estate Bd., Inc., 120 F.TC. 882 (1995); In re United Real Estate Brokers of Rockland, Ltd., 116 F.TC. 972 (1993); In re American Indus. Real Estate Ass 'n, 116 F.TC. 704 (1993); In re Puget Sound Multiple Listing Serv., 113 F.T.C. 733 (1990); In re BellnghamWhatcom County Multiple Listing Bureau, 113 F.T.C. 724 (1990); In re Metro MLS, Inc., No. C-3286, 1990 WL 10012611 (F.T.C. Apr. 18, 1990); In re Multiple Listing Servo of the Greater Michigan City Area, Inc., 106 F.T.C. 95 (1985); In re Orange County Bd. of Realtors, Inc., 106 F.TC. 88 (1985). 2 Real estate brokers have a long history of using the local MLS to restrain competition. See, Real Estate Bds., 339 U.S. 485 (1950) (board of real estate brokers conspired to fix commission rates); United States V. Realty Multi-List, Inc., 629 F.2d 1351, 1370 e.g., United States V. National Ass 'n of (5th Cir. 1980) (MLS restrictions on membership violated antitrust laws); Austin Bd. of Realtors V. E- Realty, Inc., 2000 WL 34239114, at *2 (W.D. Tex. Mar. 30,2000) (issuing preliminary injunction against realtor board's restrictive practices concering MLS access for Internet-focused realty company); Cantor v. Multiple Listing Serv., Inc, 568 F. Supp. 424, 430-31 (S.D.N.Y. 1983) (MLS bylaw restricting members' right to post non-MLS lawn signs was unrasonable restraint of trade); Oates v. Eastern Bergen County Multiple Listing Serv., Inc., 273 A.2d 795, 800 (N.J. Super. Ct. Ch. Div. 1971) (concerted action by real estate board to deny access to MLS violated the antitrust laws); Grilo V. Bd. of Realtors, -1­ using Exclusive Agency agreements. These policies restrict the ability of limited service brokers to gain important competitive benefits from the Realcomp MLS that full-service brokers enjoy. These policies therefore restrain competition from discount brokers. Competition from limited service brokers offers substantial benefits for consumers. Traditional full-service brokers offer home sellers a full array of services in return for a commission (generally 5-6% of the selling price) that includes payment for the broker working with the seller as well as the broker working with the buyer. Limited service brokers, however, "unbundle" both services and commissions. By unbundling services, these brokers allow consumers to purchase only the real estate services they want, permitting them to save money by doing some of the work themselves. By using Exclusive Agency agreements, limited service brokers unbundle their commission from the commission for the broker working with the buyer. Under an Exclusive Right to Sell agreement, the seller must pay the full commission whether or not the buyer is represented by a broker. Limited service brokers can therefore offer home sellers substantial savings if they can sell the home on their own, i. e., if they find a buyer who is not using a broker. The Realcomp MLS is critical to brokers' ability to compete. It offers two key services to its over 14,000 members. First, members can list sellers' homes for sale on a central database, exposing these listings to thousands of other Realcomp members representing potential buyers. Second, Realcomp sends its members' listing information to key public Internet websites, exposing these listings directly to milions of potential home buyers. These services give brokers a significant competitive advantage. 219 A.2d. 635, 640-648 (N.J. Super. Ct. Ch. Div. 1966) (same). -2­ Realcomp, however, took action to eliminate the competitive benefits of these services for limited service brokers while fully maintaining the benefits for full-service brokers. Realcomp first banned all Exclusive Agency listings from its feed oflisting information to public Interet sites. It reserved this key competitive advantage only for full-service Exclusive Right to Sell listings. Realcomp then changed how brokers could search for listings within the Realcomp MLS database, setting the search default to Exclusive Right to Sell listings. Finally, to ensure that brokers using Exclusive Right to Sell listings in fact provide "full servce," Realcomp imposed minimum service requirements on these tyes of listings. The result was predictable. Realcomp's "Website Policy" and "Search Function Policy" have severely impacted limited service brokers and restrained competition. Due to Realcomp's Policies, limited service brokers have exited the market, been deterred from entering the market, incurred increased costs, been forced to offer inferior products, and been forced to develop more expensive products to comply with Realcomp' s rules and gain the competitive benefits offered only to full-service brokers. Realcomp's policies also restrain competition by limiting the package of services and terms that competing brokers can offer to consumers. In effect, Realcomp members have agreed that they wil not offer consumers certain packages of services ­ Exclusive Agency listings combined with the full benefits of the Realcomp MLS. The market- wide impact of Realcomp's actions is confirmed by Realcomp's own data, which shows that limited service Realcomp's policies have caused a significant, overall reduction in the use of brokers. Realcomp's actions violate Section 5 because they uneasonably restrain competition. Realcomp's Policies are the result of a combination of competitors. They restrict the ability of -3­ brokers offering discounted, limited services to compete effectively. They also withhold from consumers a product they desire. Realcomp has market power, and through the Policies, it has used that power, resulting in an overall market reduction oflimited service brokerage activity. The Policies therefore insulate full-service brokers from price competition and harm consumers by forcing consumers to pay (and pay more) for unwanted real estate brokerage services. II. STATEMENT OF FACTS A. Real Estate Brokers and the Brokered Real Estate Transaction The vast majority of consumers use a real estate broket to sell their homes. (CCPF 144).4 Many consumers also use brokers to help purchase a home. A brokered real estate transaction therefore often involves two brokers: a "listing broker" who assists the seller; and a "cooperating broker" who assists the buyer. (CCPF 146). The listing broker markets the seller's home and may provide a number of other services, such as helping set the initial list price, negotiating with potential buyers, and assisting in the "closing" of the transaction. (CCPF 149). Listing brokers traditionally receive a commission the sale price of based on a percentage of the home. (CCPF 156). The cooperating broker brings the buyer to the transaction.5 To attract cooperating brokers with ready buyers, listing brokers (in agreement with the seller) make an "offer of compensation" for "procuring" the buyer. (CCPF 166). The offer of compensation is typically a 3 We use "real estate broker" to encompass brokers and their agents. 4 CCPF refers to Complaint Counsel's Proposed Findings. 5 Cooperating brokers may represent the buyer as a "buyer's agent" or may act as a "sub-agent" for the seller as a "selling agent." (CCPF 159-60). -4­ percentage of the sales price. (CCPF 172,351). In a brokered transaction, therefore, the seller effectively pays a commission to both the listing broker and the cooperating broker. (CCPF 167). B. The Local MLS Listing brokers market homes by listing them on the local MLS. (CCPF 466, 232 (only brokers who are members of the MLS can list properties)). The modem MLS disseminates listings to cooperating brokers through a closed database system and directly to potential buyers by posting listings on various public websites. To find homes for buyers, cooperating brokers search the MLS database. (CCPF 145, 158). Listing brokers therefore not only provide information about the home in their listings, they must also include the offer of compensation for every listing placed on the MLS. (CCPF 234,350). The information on a listing's offer of compensation, which is enforceable through binding arbitration, makes the MLS unique among sources of information about homes for sale. (CCPF 234, 237-38, 314, 362-63). The MLS database therefore facilitates cooperation between brokers by disseminating information and providing orderly rules governng broker cooperation. (CCPF 227-30). The modern local MLS also disseminates listing information to public websites. Through these public websites, home buyers can search for available properties for sale. (CCPF 240-41). These websites typically include web sites such as Realtor.com (the offcial consumer website of Realtors ("NAR")), public websites operated by the local MLS, and the National Association of member broker and agent websites, known as "IDX''¡ websites. (CCPF 243-47). 6 IDX (Internet Data Exchange) allows MLS member brokers to post on their own website listings from the MLS from all other brokers that participate in the mx. (CCPF 245-47). -5­ C. Realcomp Realcomp operates the largest MLS in Michigan.7 (CCPF 282). It has over 2,200 real estate office members and approximately 14,000 members,8 who "compete with one another to provide residential real estate brokerage services to consumers." (CX 32-002 (Answer~ 4); CCPF 278-79, 282). 1. Realcomp's Governance and Membership Realcomp is owned and controlled by competing real estate brokers. It is owned by seven local realtor boards and associations ("Shareholder Boards"), which are local associations of real estate professionals affiiated with NAR. (CCPF 253-56). The seven Shareholder Boards are comprised of competing real estate brokers. (CCPF 255). Realcomp is controlled by a Board of Governors, which sets Realcomp' s rules and policies. (CCPF 258, 268). The Board is comprised of representatives from the Shareholder Boards, who are competing real estate brokers. (CCPF 257,263,265; CX 59-011 (Realcomp bylaws require that at least one of the Governors from each Shareholder Board must be "actively practicing real estate")). 2. The Realcomp MLS Member Services Realcomp provides a host of services to its members. (CCPF 303, 368, 413-14, 432-36, 437,440). Two are most important. First, Realcomp operates a traditional MLS database, which 7 The Realcomp MLS covers Southeastern Michigan, i.e., Oakland, Wayne, Livingston, and Macomb counties. 8 Realcomp's membership is open to any real estate broker who is a member of one of the Shareholder Boards. (CCPF 286). Thus, any Michigan licensed real estate broker can join NAR and one of the Shareholder Boards, and in tumjoin Realcomp. (CCPF 286). -6­ allows its members to exchange listings and make offers of compensation. (CCPF 303, 313-15). The Realcomp MLS database allows listing brokers to expose their listings to thousands of Realcomp cooperating brokers. (CCPF 229, 279, 465, 473). This gives Realcomp cooperating brokers the opportnity to bring a buyer to the sale and earn a commission. But, as Realcomp has stipulated, Realcomp does not require that a Realcomp cooperating broker (or any cooperating broker) paricipate in a transaction facilitated by its MLS.9 (JX 1-05; CCPF 322). Properties listed on the Realcomp MLS therefore may be sold to buyers represented by Realcomp members, represented by brokers who do not belong to Realcomp, or not represented by any broker. (CCPF 322). In fact, Realcomp members testified they do sell listed propertes to unrepresented buyers; it "happens all the time in open house(s)." (CX 413 (Kersten, Dep. at 45­ 46); CCPF 173). Second, Realcomp feeds listing information to a network of public real estate websites, giving these listings wide exposure directly to potential buyers. These "Approved Websites" fall into four categories: · Realtor.com, the most popular real estate website in the country; · MoveInichigan, Realcomp's own public MLS website, which is also the exclusive provider of real estate listing information for ClickOnetroit, the most popular website in Southeastern Michigan; · Realcomp brokerage firm websites through the Realcomp IDX; and · Realcomp agent websites, also through the IDX. (CCPF 369, 405, 407). 9 Another benefit Realcomp gives to its members is data-sharing with nearby MLSs, which exposes Realcomp listings to cooperating brokers who are not members ofRealcomp. (CCPF 413-16, 421). -7­ This allows Realcomp listing brokers to gain wide exposure for their listings. io The Realcomp IDX, for instance, includes _ of local broker and agent websites, such as sites forlocal franchises ofmajorreal estate firms (e.g., Century 21, ReMax, Keller Wiliams, Coldwell Banker, etc.) that are sometimes integrated into the franchisor's national website. (CCPF 412,666,644). In addition, this information is provided to a local cable television provider for the Home Preview ChanneL. (CCPF 435-36). As one Realcomp document puts it, through the Realcomp MLS, listing brokers can reach: 1) Approximately 15,000 Realcomp II Ltd. MLS Subscribing REALTORS. 2) Milions ofInternet users shopping for homes on MoveInichigan.com, REALTOR. COM, and Realcomp Subscribing Brokers' IDX (Internet Data Exchange) websites. 3) Over 1,250,000 cable TV viewers in approximately 350,000 households subscribing to Comcast's Digital Cable-TV in Southeastern Michigan. (CX 272; CCPF 601). As with Realcomp's MLS, Realcomp has stipulated that transactions facilitated by Realcomp's feed to the Approved Websites need not involve a Realcomp cooperating broker (or any cooperating broker). (JX 1-05; CCPF 374). These properties maybe sold to buyers represented by Realcomp members or to any other buyers. (CCPF 322,374). D. Real Estate Brokerage Service Models and Listing Contracts Real estate brokers compete to obtain listings, i.e., brokers compete to sell their services to home sellers. (CCPF 204). The form of competition offered by a broker - e.g., full service 10 Realcomp's feed oflisting informtion also automatically includes any updates to the listing that are entered on the MLS. (CCPF 600). -8­ versus limited service - is largely reflected by the type of "listing agreement" used by the broker. A listing agreement spells out the nature of the relationship between the broker and the seller, and includes the compensation to be paid to the listing broker, the offer of compensation to be made to cooperating brokers, and the conditions under which those commissions must be paid. (CCPF 151, 153, 155). 1. Traditional Full Service Brokers and Exclusive Right to Sell Listings Traditional full-service brokers provide a range of services to sellers, including listing the propert on the MLS, holding open houses, showing the propert to potential buyers, and assisting with the closing of the transaction. (CCPF 149, 180,333). These services are want but instead must "bundled," i.e., sellers cannot pick and choose which services they purchase the whole package. These brokers tyically charge between 5-6% of the sale price. (CCPF 180). To offer their services, full-service brokers use "Exclusive Right to Sell" listing agreements. (CCPF 189,328). As Realcomp has admitted, this "is the form oflisting agreement traditionally used by listing brokers to provide full-service residential real estate brokerage services." (CX 32-003-004 (Answer iì 8); CCPF 328). This form of listing agreement has important implications for the seller. An Exclusive Right to Sell listing requires the seller to pay the listing broker the full commission regardless of who finds the buyer. (CCPF 176-79). In other words, the seller pays the full commission (the listing commission and the offer of compensation) whether or not a cooperating broker brings the buyer. The Exclusive Right to Sell listing thus bundles the listing broker's commission and the offer of compensation. (CCPF 176-79, 202). -9­ For example, with an Exclusive Right to Sell listing that calls for a 6% commission and an offer of compensation of 3%, if a cooperating broker brings a buyer, the seller pays the 6% commission, and the listing broker keeps a 3% commission and pays the cooperating broker the 3 % offer of compensation. But if there is no cooperating broker, the seller stil pays the full 6% commission and the listing broker wil keep both the 3% commission and the 3% offer of compensation because there is no cooperating broker to pay. (CCPF 172, 178, 180, 190) (brokers in Southeastern Michigan often charge 6% listing commissions, and over. of Realcomp listings offer a 3% offer of compensation)). 2. Limited Service Brokers and Exclusive Agency Listings Limited service brokers offer sellers a less expensive alternative. Instead of providing a full bundle of services, these brokers allow consumers to select from a menu of unbundled brokerage services and to "self supply" the other services. (CCPF 191-92). Limited service brokers therefore provide a middle ground for consumers between sellng completely on their own ("for sale by owner" (FSBO)) and traditional full-service brokers who provide all of the services at a signficant cost. (CCPF 193). As one industry study put it, limited service brokers meet a "consumer demand for lower cost brokerage services where consumers are wiling to carr out some of the homeselling tasks themselves that otherwise would be performed by real estate professionals." (CX 533-041; CCPF 194). By using Exclusive Agency listing agreements, ii limited service brokers also unbundle broker commissions. (CCPF 187,202-03,341). An Exclusive Agency listing reserves to the 11 We refer to all "non-ERTS" listings - Exclusive Agency, Limited Service, and MLS Entry Only - as Exclusive Agency listings. -10­ seller a right to sell their home on their own, in which case the listing broker is paid a reduced or no commission when the home is sold. (CCPF 183). Thus, Exclusive Agency listings allow sellers to avoid paying for the services of a cooperating broker (through the offer of compensation) when an unrepresented buyer purchases propert. (CCPF 184-87,203). For example, assuming an EA listing calls for the payment of an up-front $499 flat fee to the listing broker and a 3% offer of compensation: if a broker brings a buyer, the seller pays the up- front fee and the offer of compensation, but if the buyer went directly to the seller and there was no other broker involved, then the seller wil have paid the up-front $499 flat fee, but would not owe any other additional commission. (CCPF 185). The potential savings can be substantiaL. For a $300,000 home, for instance: Cooperating Broker ERTS No Cooperating Broker Full commission (total: $18,000) (listing broker 3%; cooperating broker 3%) Commission or fee to listing broker ($499) + offer of compensation (3%) to cooperating broker (total: $9,499) Full commission (total: $18,000) (listing broker 6%) Commission or fee to listing broker ($499) No offer of compensation paid (total: $499) EA 3. Competition Among Full Service and Limited Service Brokers Full-service and limited service brokers compete head-to-head. (CCPF 206). Limited service brokers are a fairly new form of competition that grew in importance in the early 2000s. (CCPF 214). And seller awareness of these brokers has been increasing - to _. (CCPF 224; Murray, Tr. 174-175 ("if more sellers are aware that there are alternatives that fits them.")). that are lower cost, the more sellers are going to at least investigate it and see if Thus, as one industr white paper put it, limited service brokers have become "a signficant competitor to full-service brokerages." (CX 375-029; CCPF 215). -11­ This competition has put price pressure on full-service broker commissions. (CCPF 221­ 26). As Mr. Murray explained at trial, "Anytime you have a competitor in the market who is offering perhaps a little bit fewer services and lower price to some degree, it's going to induce some competition on price into the marketplace." (Muray, Tr. 174; CCPF 223). A NAR study similarly recognized, "Online brokerage models or low-service market discounters wil put continuing pressure on broker and agent commissions." (CX 403-009; CCPF 221). Mr. Whitehouse, a full-service broker in Southeastern Michigan and former President of the Michigan Association of Realtors, admitted this to be tre: Q: So you would agree that full service brokers are in competition with discount brokers, right? A: Absolutely. We're in competition with eveiy brokerage. Q: And they have to find ways to resist the pressure from discount brokers, right? A: I would say so, yes. the things discount brokers are putting pressure on is the commission Q: One of rates, right? A: Um-hum. Yes. (CX 421 (Whitehouse, Dep. at 25); CCPF 226). Limited service brokers put price pressure on full-service brokers in strong seller's markets as well as in strong buyer's markets, such as the curent climate in Southeastern Michigan. (CCPF 216). This is because they appeal to sellers who may not have much equity in their homes. (CCPF 216-18). As Mr. Murray explained at trial, sellers in this position are "going to look for the lowest-cost alternative they can to sell their home because, whatever it is, they're going to have to write a check to get out of their house." (Murray, Tr. 169-171 (noting that these sellers are "price-conscious shoppers"); CCPF 195, 217). -12­ E. Realcomp's MLS Member Services are Signifcant to Brokers' Abilty to Compete 1. "Exposure" of Listings is Important to Consumers and Critical to a Broker's Abilty to Compete To compete effectively, brokers need exposure for their listings - one of the "keys" to sellng a home. (CCPF 454; CX 352-001). Exposure maximizes the chances that a broker wil be able to match a wiling buyer with a wiling seller. (CCPF 454-55; CX 352-001). More specifically, exposure is "critical" to a broker's ability to compete because consumers demand it. . (CCPF 456-57; Muray, Tr. 183 ("(e)xposure is critical" in the real estate industr); CX 405 (Baczkowski, Dep. at 38-39 (noting that home sellers "want their propert exposed to as many people as possible. . . ."). This demand is not uneasonable. As Realcomp Governor Robert Gleason explained, less exposure: means less price, more marketing time, more expenses involved, lower price on your home, more days on the market, more carrng costs; in other words, it's more expensive for everybody concerned. (CX 38 (Gleason, Dep. at 123-124); CCPF 458). 2. Exposure Through the Realcomp MLS Database is Important For Brokers to Compete Effectively The Realcomp MLS gives this sort of exposure and is therefore critical for brokers to be able to compete effectively in Southeastern Michigan. (CCPF 473-77, 502-04; Murray, Tr. 179­ 180; G. Moody, Tr. 870 ("The MLS is critical to success, especially in Michigan.")). For brokers in Oakland, Wayne, Livingston, and Macomb counties, no other MLS provides the geographic reach or membership size ofRealcomp. (CCPF 513). -13­ a. The Importance of the Local MLS The ability to list properties on the local MLS is vital for brokèrs to compete effectively. (CCPF 463-93). As one Realcomp member advises consumers shopping for a broker, "An absolute must is that the (broker) subscribes to the local computerized Multiple Listing Service (MLS) so that your propert is exposed to the maximum number of potential buyers." (CX 307­ 001; CCPF 511); see also Federal Trade Commission and u.s. Departent of Justice, Competition in the Real Estate Brokerage Industry at 12 (April 2007) ("MLSs are so important to the operation of real estate markets that as a practical matter, any broker who wishes to compete effectively in a market must participate in the local MLS."). MLSs expose listings to all other MLS members, "drmatically increasing" the listing brokers' marketing reach. (RX-154-A-026-027; CCPF 475,516; Sweeney, Tr. 1315 (the MLS provides "a huge buyer stream available" for brokers' listings)). The MLS provides "(o)ne of the most effective networks" available to reach potential buyers, and, as one Realcomp Governor admitted, not listing on the MLS "would be like tyng my hands behind my back." (CX 40 (E1ya, Dep. at 35-36); CX 525 (Adams, Dep. at 76-77); CCPF 470, 474). Cooperating brokers search the local MLS because it is a "single comprehensive source of information about listings from other brokers in the area." (Sweeney, Tr. 1343; CCPF 469). Access to "the most inventory possible" in the MLS is important because "eveiy buyer's needs are somewhat different, just like every home is a little different." (Murray, Tr. 181-182; CCPF 467). As one Realcomp Governor has admitted, a cooperating broker not using the MLS "would be very ineffcient." (CX 42 (Nead, Dep. at 133); CCPF 468). il fact, listings on the MLS are -14­ particularly attractive to cooperating brokers because each listing carries with it an enforceable offer of compensation. (CCPF 237, 360, 362-63). The significance of the MLS is evident from the fact that approximately 88% of sellers using brokers reported in 2006 that their home was marketed on the MLS - more than any other marketing tool available to brokers. (CCPF 486,573). Realcomp Govemors and members repeatedly testified that they could not recall a client ever askig for their listing to not be placed on the MLS. (CCPF 478-79). Because of the MLS 's valuable exposure, the ability to market a home on the local MLS impacts the ability of a listing broker to compete for listings with potential new clients. (CCPF 487,491). As the industry expert, Mr. Murray, confirmed, a listing broker "whose properties were not displayed on an MLS would be at a significant competitive disadvantage to those brokers whose properties were listed on the MLS." (RX 154-A-32; CCPF 490). Partcipating in the local MLS "absolutely" impacts a broker's ability to obtain new listing agreements because sellers "require it or demand it," and an inability to list on the MLS wil be highlighted by other listing brokers who belong to the local MLS and are competing for that listing.12 (Murray Tr. 186-188; CCPF 490, 487). 12 Notably, however, brokers wil avoid participating in two or more MLS services "ifthey can help it." (Murray, Tr. 184; CCPF 494). As explained by Mr. Sweeney at trial, the monthly fees of belonging to two MLSs are a "significant cost only to be incurred if necessary," and "a bigger cost is the administrative hassle of entering the listings in both systems." (Sweeney, Tr. 13 12, 1340; CCPF 496­ 497). The costs of participating in more than one MLS include: "the payment of multiple MLS user and perform multiple searches over the same geographic area on behalf of office fees; duplication of efforts in ters of data entry, system access and training sessions; having to buyers; learng different terms and terminology; and following multiple policies, rules and data display requirements." (RX 154-031­ 032; CCPF 495). -15­ b. Access to the Realcomp MLS Is Important For Brokers in Southeastern Michigan To Compete Effectively Realcomp is the local MLS for brokers in Southeastern Michigan. (CCPF 512, 515-16). It is the largest MLS in Michigan. (CCPF 282). And access to the Realcomp MLS is important for brokers to compete effectively. (CCPF 502-04, 513-35). Numerous brokers testified that Realcomp is their local MLS. (CCPF 512). By placing their listings on the Realcomp MLS, brokers expose their listings to the thousands ofRealcomp members working with buyers, which is "great exposure for a house that's for sale." (Mincy, Tr. 318; CCPF 505). As one owner of a local brokerage explained, the Realcomp MLS is "critical" to doing business as a broker in Southeastern Michigan because: for southeast Michigan, Realcomp is the MLS, and that's where all the realtors go to find the houses, and what they do is they search the MLS for their buyers' criteria, and so this is where all the realtors go to find out what's available in the market for sale. (G. Moody, Tr. 856-857; CCPF 502). As Realcomp's own documents state, Realcomp's size­ 2,230 participating offces and "nearly 15,000" participating agents - allows brokers to "(mJake more sales through co-op arrangements with nearly one-half of all REAL TORS in Michigan." (CX 224-001; CCPF 504). These "cooperative opportnities," according to Realcomp, "are priceless." (CX 222-007; CCPF 504). Although there are MLSs in the areas adjacent to Realcomp, none are an "effective alternative(J that provide the geographic coverage or membership size that is offered by Realcomp.,,13 (RX 154-A-032; CCPF 513). Realcomp's size as the largest MLS in Southeastern 13 For example, one Realcomp Governor explained that "(i)t never made sense" to list properties in the An Arbor or Flint MLSs because his local MLS is Realcomp; in fact, he does not even search the MiRealSource MLS on behalf of his buyers because MiRealSource typically services an area other than -16­ Michigan (and in the entire state of Michigan) is significant because an MLS with more listings and more members provides a greater benefit to brokers than do smaller MLSs. (CCPF 517-519; RX 154- A -031 ("The value of an MLS increases with the more listings it has because that increases the likelihood that brokers wil be able to match a wiling buyer with a wiling seller.")). Mr. Murray explained the benefits of a larger MLS: you're a listing broker, you've got more potential cooperating brokers with more buyers to help sell your home. If you're a cooperating broker, you've got more inventory to look at. If you're working with a buyer, I mean, it would be as if - if I were a cooperating broker in those four counties and there were another MLS with only 3,000 participants, well, where would I list my home? I would list it on the bigger one, because there's more cooperating brokers, more people and chances to get my client's home sold. if (Muray, Tr. 182-183, 189 (noting that Realcomp's market shares are "sufficiently high" in Wayne, Oakland, Livingston, and Macomb counties that brokers need access to the Realcomp MLS to compete effectively); CCPF 518-19). 3. Exposure Through Realcomp's Feed to the Approved Web sites is Important sites is important for brokers For Brokers to Compete Effectively Realcomp's feed of listing information to the Approved Web to compete effectively. (CCPF 536, 673). Buyers use the Internet - and, in paricular, the four categories of web sites covered by the Approved Web sites - as an "integral" par of their home search. (CCPF 537, 594-97, 599). Marketing listings on those websites is therefore important to expose a broker's listings to potential buyers as well as to compete for new listings. (CCPF 536, 673-76). where his buyers are looking to purchase a home. (CX 39 (Taylor, Dep. at 15-17,44-45); CCPF 512). -17­ a. Internet Marketing is Necessary to Reach Buyers Searching for Homes on the Internet Industry studies confirm that the Internet is now when buying a home CCPF 540). According to the 2006 NAR Profie of in 2002. (CX 535-006, 025-026, in camera; Home Buyer and Sellers, 80% of all home buyers use the Internet in their home search - a percertage that has increased steadily over the past decade, from 2% in 1995 and 41 % in 2001. (CCPF 541). Byway of comparison, buyers used the Internet more than any other information source except brokers (85%) - more than yard signs, print newspaper ads, open houses, and home magazines. (CCPF 542). These consumer surveys are confirmed by website statistics that show consumer usage of real estate websites has skyrocketed over the past five years. (CCPF 549-51). Between December 2002 and December 2006, the number of visitors to real estate websites grew from 21.44 milion to 31.23 milion, and the amount of time consumers spent on real estate websites grew from 294.2 million to 1.259 bilion minutes. (CCPF 551). As a result of searching the Internet, over 70% of buyers went out and looked at homes they found in their searches, and 24% of buyers purchased the home they first found on the Internet. (CX 373-039-040; CCPF 553-54). By way of comparison, 36% of buyers first learned about the home they ultimately bought from their broker - down from 50% in 1996 - and only 15% of buyers reported first learning about the home they purchased from a yard sign. (CCPF buyers now come to their first meeting with a broker with a list of 555). In fact, many homes they would like to see based on their Internet search. (CCPF 536, 562). -18­ Surey data collected by NAR from buyers in Southeastern Michigan as part of their annual Profile of Home Buyers and Sellers are consistent with these national trends: Buyers Who Used the Internet Buyers Who First Found Home They Purchased on Internet in Their Home Search 10 10 30 - - ..-' 70% - 20 10% 5D 01 30 20% 10% -~ ~ 20 20 205 -.Natnal .-Realomp A.. 0% 2Ø 20 - 2_ 0% 2O (CCPF 543, 556). Notably, buyers are not using the Internet as a substitute for using a broker. (CCPF 575). To the contrar, studies show that buyers who use the Internet are more likely to use a broker than those who do not. (CCPF 576-78; Muray, Tr. 214 (testifying that this data is a "powerful statement about the fact that the Internet is not reducing the use of real estate brokerage companies for buying and sellng of homes")). Data from buyers in Southeastern Michigan are consistent with the national statistics. (CCPF 578). b. Internet Marketing is an Effective Means to Generate Leads Given the popularity of the Internet with buyers, exposing listings on the Interet is becoming one of the most effective ways for brokers to generate leads (buyers who express an interest in a propert). (CCPF 561, 569). Studies have shown that the number ofleads generated by brokerage firm websites - which does not take into account leads generated from other Internet marketing- has been growing. (CCPF 564-65, 567-68). For example, there has been approximately a 40% growth in the number of firms who attrbuted more than 25% of their leads -19­ to their website from 2004 to 2006. (CCPF 564-65). In addition, the median brokerage firm derives 7% of its actual sales from leads generated by the firm's website, which is considered a "big chun of business" to be derived from one marketing channeL. (Murray, Tr. 218-219; CCPF 566). Industry publications have confirmed the effectiveness of marketing homes on the Internet. As a 2006 paper by NAR explains, "The brokerage firm of the futue wil need to embrace the realities of the new world order and learn to convert internet leads to paying customers in order to compete effectively." (CX 380-008; CCPF 558). The largest real estate brokerage company in the United States similarly explains in a white paper that "( c )onsumers have made it a business imperative" for brokers to be more effective at Internet marketing, which is "emerging as one of the most valuable customer lead generation tools available for Realtors today." (CX 617-003-004,007; CCPF 559-561; see also CX 621-007). Brokers recognize the effectiveness of this marketing tool. (CCPF 573-574). Except for the MLS, marketing homes on the Internet is the most used marketing tool by real estate brokers. (CCPF 573-574). Specifically, 85% of sellers reported in 2006 that their home was marketed on the Internet - an increase from 73% in 2004. (CCPF 573-574; ex 373-079 (showing only 78% of sellers reported that their broker marketed their home with a yard sign)). Brokers in Southeastern Michigan are no different. They repeatedly testified that they wanted to have their listings on public websites to expose them to potential buyers. (CCPF 581, 584). The current President of Realcomp testified that the Internet has become an increasingly important avenue for marketing homes. (CCPF 583). And numerous Realcomp members -20­ advertise their ability to provide Interet marketing to potential home sellers. (CCPF 587; see also CCPF 586 (noting that consumers are also driving this trend in Southeastern Michigan)). c. Buyers Search for Homes on Four Key Categories of Web sites sites where buyers are going. web sites more than any To be effective, brokers must get their listings to the web Studies have repeatedly found that buyers visit four categories of others: MLS websites; Realtor.com; and the websites of real estate companies and real estate agents, also referred to as "IDX websites." (CCPF 593-94; CX 373-046 (finding that 40-50% of buyers reported visiting these categories of websites)). NAR found these same results in each of its annual Profie of Home Buyers and Sellers since 2004. (CCPF 594). Another 2006 industr study . (CCPF 596; CX 535-007,029, in camera_ ). By way of comparison, newspaper websites were searched by only 14% of buyers (as compared to the 40-50% of buyers visiting each of the top four categories); home magazine websites were searched by only 6% of buyers; and all other buyers. (CCPF 595). Survey data websites were collectively searched by a total of 10% of collected by NAR from buyers in Southeastern Michigan are also consistent with these national statistics. (CCPF 597). Website statistics confirm the results of the industr studies. For example, Realcomp's MLS website, MoveInichigan averaged approximately 748,000 hits each month from January 2006 to October 2006. (CCPF 631). Realtor.com, a national website, had over 3.9 milion unique users in December 2006, spending approximately 119 milion minutes on the website. -21­ (CCPF 637). IDX websites collectively had 5 milion unque visitors in December 2006, spending 103 milion minutes on the websites. (CCPF 644). This shows the "large and growing" competitive "signficance" ofIDX websites. (CCPF 644, 645). Unique Visitors to LOX Websites Total Minutes Spent on LOX Websites Unique Visitors (in Millons) 6 Total Minutes (in Millons) 120 100 80 5 4 3 2 1 60 40 20 0 0 2002 2003 2004 2005 2006 2002 2003 2004 2005 2006 (CCPF 645, 646; RX 154-A-062-063; CX 609-001,016). d. The Competitive Significance of Realcomp's Feed of Listing Information to the Approved Websites Rea1comp's feed to the Approved Websites is competitively significant because it goes to each of the four categories of key web sites most visited by buyers. (CCPF 599, 673-676). This is importt because "Internet marketing is a signficant competitive advantage to brokers - but only if they are able to advertise their propert listings on the websites visited by a signficant potential buyers in their geographic area." (RX 154-A-046; CCPF 592). As sites number of explained by Mr. Muray, it is "critical" for a broker to target his listings on the Internet web where consumers in a broker's service area are most likely to be lookig for homes. (Muray, Tr. 221-222; CCPF 593). Thus, while there are "tens of thousands of real estate Web sites. . . . and -22­ it's okay to be on some of those, but the ones you really have to be on to compete effectively are the four major sites where 40to 50 percent of all the buyers are going:" (Muray, Tr. 238 (referring to MLS websites, Realtor.com, broker and agent websites); CCPF 673). This is important not only from an exposure point of view, but also because "sellers want their information at the site that is going to best market them, and best attract the consumer." (CX 405 (Baczkowski, Dep. at 38-39, 46 (testifyng that home owners want their listings "to be at the best site possible"); CCPF 592, 676)). Thus, participating in Realcomp's feed oflisting information is important for a broker to be able to compete for new listings with potential clients looking to sell their home. (CCPF 676). F. Realcomp Adopted Policies That Limit the Exposure of Exc1usive Agency Listings As brokers began to offer discounted, limited services to consumers in Southeastern Michigan through Exclusive Agency listings, Realcomp (under the control of competing fullservice brokers) reacted by imposing rules that impair the effectiveness of these tyes of listings. Advised by more than one legal counsel that an outright ban of Exclusive Agency listings would be problematic, Realcomp tued to the next best solution - excluding Exclusive Agency listings from Realcomp's feed to public and IDX websites, and hampering the exposure of these listings within the MLS. (CCPF 765, 771, 776, 820). In June 2001, just as limited service brokerage services began to grow throughout the real estate industr, the Realcomp Board of Governors voted to exclude Exclusive Agency listings from "the data that is sent to the real estate Internet advertisers." (CX 2-003; CCPF 774). In September 2001, after the Board "reviewed a memorandum from Legal Counsel" on the issue, it again voted to exclude Exclusive Agency listings "from all data extracts to the Internet real estate -23­ Web sites publishing Realcomp data." (CX 3-002; CCPF 776). Realcomp implemented these changes in October 2001. (CCPF 778). The "Website Policy" was thus born. (CCPF 766). Under the Website Policy, Exclusive Agency listings are banned from Realcomp's feed of listings to Realtor.com, MoveInichigan, ClickOnDetroit, and the Realcomp IDX. (CCPF 766, 780-783). Realcomp's Rules and Regulations reflect the Board's decisions. As of July 2003, those rules stated: Exclusive Agency, Limited Service and MLS Entr Only Listings wil not be distributed to any Real Estate Internet advertising sites. (CX 4-012; see also CCPF 822 (current rules)). With regard to the Realcomp IDX, the rules stated: The "IDX Database" is the current aggregate compilation of all active, full service. exclusive right to sell listings of all IDX (participants), except those listings where the propert seller has opted out ofInternet publication by so indicating on the Exclusive Right to Sell Contract. (CX 4-021 (emphasis added); see also CCPF 826 (curent rules)). In August 2002, the Realcomp Board of Govemors considered excluding Exclusive Agency listings altogether. (CCPF 784; CX10-003). Apparently because of a "requirement from NAR to allow these types oflistings into the MLS," the Board chose not to do so. (CXlO-003; CCPF 785-787,819). Instead, it told the Realcomp staff to "revisit the current policies and procedures regarding flagging" Exclusive Agency listings and the "fine for failure to comply." (CX 10-003-004; CCPF 788-789). Realcomp later revised its Policy Handbook to require that Exclusive Agency listings be indicated with the "proper flag." (CX 5-007; CCPF 790-791). In August 2003, the Board was again confronted with the issue of how to deal with Exclusive Agency listings. MiRealSource, a nearby MLS that is not associated with NAR, had -24­ decided to exclude these types of listings. (CCPF 792; CX 9-003). In response, the Board voted to "expedite the enhancement of defaulting all searches to include only Exclusive Right to Sell Listings and that the other listing tyes are shown only by request." (CX 9-003; CCPF 793-796). And so the "Search Function Policy" was born. To fully implement the Search Function Policy and to make the Website Policy more enforceable, Realcomp needed to ensure that the listing type was indicated on every listing in the MLS. In late 2003, Realcomp informed its membership of its plans, explaining that this change was precipitated by the entry of limited service brokers - or as Realcomp described it, "new levels of service being offered by MLS Subscribers." (CX 18-003; CCPF 806). Realcomp therefore made listing tye a mandatory entr for all MLS listings. (CCPF 806-807). Realcomp members could label their listings as an Exclusive Right to Sell (ERTS) ("a listing where the listing agent offers full service," i.e., provides five minimum services defined by Realcomp), an Exclusive Agency (EA), a Limited Service (LS), or an MLS Entr Only (MEO) listing. (CX 18­ 003-004 (emphasis added); CCPF 807-808, 333). Mandatoiy labeling allowed Realcomp to implement the Search Function Policy. Within the Realcomp MLS database, the search fuction automatically defaults to Exclusive Right to Sell listings. (CX 18-005; CCPF 794-796). Mandatory labeling also allows Realcomp to more fully enforce its Website Policy. The labels allow Realcomp to completely filter out Exclusive Agency listings from the feed to public web sites and from the IDX. (CCPF 791,811). Realcomp has also adopted strict fines to penalize any members who do not accurately represent the nature of their listing agreement. As of 2004, the fine for a first offense is $250, a second offense in a 12 month period is $1,000, and -25­ a third offense $2,500. A fourh offense wil result in a 45 day suspension from the MLS for the entire brokerage offce. A fifth offense wil result in dismissal from the MLS. (CX 6-014; CX 7­ 015; CCPF 813). Despite these efforts, Realcomp's traditional full-service brokers stil peppered Realcomp with the question of why Realcomp continued to allow Exclusive Agency listings on the 2004, Realcomp's CEO broadcast to members the reasons: Realcomp MLS. In April REALTORS requires MLSs to listing tyes (i.e. Exclusive Right to Sell (Full Service), Exclusive Agency, Limited Service, and MLS Entry Only). . . . The first reason is that the National Association of accept all In addition, Realcomp has been advised from more than one legal counsel to accept and include these listings. (CX 29; CCPF 819-20). Ms. Kage, however, promised to "continue to carefully monitor this controversial and sometimes confusing topic." (CX 29). True to her word, after attending a meeting with 45 other MLSs in July 2004, Ms. Kage reported to Rea1comp's members: I spoke with several more MLSs during these meetings to determine whether any of them have adopted rules that would prohibit listings that are not 'Full Service/Exclusive Right to Sell' from being included in their database. None of those I spoke with had made any changes to their MLS rues to prohibit these types of listings. (CX 28). Consistent with Ms. Kage's finding, the evidence in this case shows that other MLSs have no problem with sending Exclusive Agency listings to public websites. (CCPF 1249). G. The Impact of Realcomp's Policies Though only second best, Realcomp's Policies have been effective. The evidence shows that the Policies hobble one ofthe most important services that brokers can offer to consumers: wide exposure of the consumers' listings. (CCPF 868-940). Not surprisingly, the evidence, -26­ including uniform testimony from discount brokers, shows that the Policies affect limited service brokers' ability to compete. (CCPF 941-1068). Data from Realcomp (as well as ten other MLSs for comparison puroses) bear this out: the market share of Exclusive Agency listings in the Realcomp MLS has dropped significantly due to the Realcomp Policies. (CCPF 1069-1122). Realcomp's Policies eliminate a product from the market - an Exclusive Agency listing with full exposure to the public through the Realcomp Approved Websites and unimpeded exposure to cooperating brokers searching the Realcomp MLS database. (CCPF 861, 867, 1124, 1200). This is the product that limited service brokers have been seekig to offer, and this is the product that the evidence shows consumers desire. (CCPF 1156-72). Due to Realcomp's Policies, consumers are left to choose either a full service, Exclusive Right to Sell listing or a degraded version of an Exclusive Agency listing. (CCPF 870,1029, 1055). 1. The Realcomp Policies Limit the Exposure of Exclusive Agency Listings to Buyers and Cooperating Brokers The Realcomp Policies impact both of the channels that limited service brokers use to gain wide exposure for Exclusive Agency listings. The Website Policy prevents discount brokers from being able to expose these listings directly to buyers on key Internet sites. The Search Function Policy limited the exposure of these listings to cooperating brokers. a. The Realcomp Website Policy Severely Lioots the Exposure of Exclusive Agency Listings to Buyers The Website Policy bars Exclusive Agency listings from Realcomp's feed to the four most popular categories of real estate websites. (CCPF 599, 861). This severely limits the exposure of Exclusive Agency listings to buyers. (CCPF 868). -27­ The Website Policy effectively excludes Exclusive Agency listings from three of the top four categories of websites used by buyers - MLS web sites, real estate company sites, and real estate agent sites. (CCPF 594-96, 599). Discount brokers have no alternative means to place their Exclusive Agency listings on MovelnMichigan, the local MLS website. (CCPF 875). MoveInMichigan is the exclusive provider of real estate listings for ClickOnDetroit; thus, Exclusive Agency listings also do not receive exposure through that popular website either. (CCPF 876). Nor do discount brokers have any practicable means to put Exclusive Agency listings onto Realcomp IDX sites (real estate company and real estate agent sites).14 (CCPF 878­ 80). Limited service brokers have been able to bypass Realcomp's Website Policy to reach Realtor.com, but only by expending additional time and money "double listing." (CCPF 881­ 85). As Realcomp's own trial witness admitted, double listing imposes a "significant cost only to be incurred if necessary" both in dollar terms and administrative burden: It's not just the double entr, on the entry, it's the maintenance, every time there's a price change, you have to do it in two systems, any time there's any change whatsoever at least reported in the system, you have to do it twice. Yes, that is a burden. An administrative burden. (Sweeney, Tr. 1340; CCPF 883). Access to Realtor.com, however, does not overcome the competitive disadvantage created by Realcomp's Website Policy. The evidence shows that MoveInMichigan and the Realcomp IDX sites are key means to expose properties to buyers in Southeastern Michigan. (CCPF 874, 632-36,649-55). Further, the characteristics of the real estate market amplify the impact of 14 Realcomp also does not send Exclusive Agency listings to the Home Preview ChanneL. (CCPF 824). -28­ barring Exclusive Agency listings from these popular websites. (CCPF 890-98). Because buyers have very particular needs and tastes and homes are unique, there may be very few buyers who would truly be interested in purchasing a paricular home. (CCPF 890). This means that wide exposure of a listing on the most popular websites is criticaL. (CCPF 893-98). Preventing Exclusive Agency listings from reaching those key websites can therefore have severe effects ­ reducing the number of bidders on a home, which reduces the price. (CCPF 893-94). Other websites - the Trulias and Googles of the world - simply are not good substitutes for these Realcomp Approved Websites. (CCPF 901-07). They are not as popular, many charge posting fees, others charge referral fees, and many simply do not have a presence in Southeastern Michigan. (CCPF 901). Moreover, unlike the automatically updated listings being fed from the Realcomp MLS, a broker must update listings on each of these other websites separately. (CCPF 600, 907). As the testimony at trial shows, other websites are not realistic alternatives for limited services brokers. (CCPF 901-07). As Mr. Muray explained at trial, the Website Policy restricts the ability of discount brokers to compete: The Realcomp Web site policy makes it either impossible or very diffcult for a broker offering exclusive agency, limited service or MLS entry to reach buyers at the same level that their competitors offering exclusive right to sell reached them. What it means is that they wil not be as effective at their marketing. It means likely they wil not get as many leads. . . . It wil mean that they are not as competitive in getting listings, because if they can't promise viewership on all these (websites) and their competitors can, that's going to be meaningful to someone looking for a listing broker. So they are greatly disadvantaged by this Web site policy. (Muray, Tr. 245-246; CCPF 873). -29­ b. The Realcomp Search Function Policy Limited Exposure of Exclusive Agency Listig to Cooperating Brokers The Search Function Policy is aimed at the other critical means of marketing listings ­ exposing them to cooperating brokers through the MLS database. While the search function default can be overridden by a click of a button, the evidence shows that it nonetheless limited the exposure of Exclusive Agency listing. (CCPF 908-40). Realcomp's own data tell the story. The data show that cooperating brokers viewed Exclusive Agency listings in the Realcomp MLS database far less than they viewed Exclusive Right to Sell listings. (CCPF 911-16; CX 228-007 (Realcomp interrogatory response showing Exclusive Right to Sell listings viewed on average 201 times per month versus 94 times per month for Exclusive Agency listings)). The data also show that cooperating brokers emailed Exclusive Agency listings to their clients far less than Exclusive Right to Sell listings. (CCPF 917-20). According to Realcomp's own calculations, brokers emailed Exclusive Right to Sell listings on average 286 times per month while they emailed Exclusive Agency listings on average only 1 time per month. (CCPF 919). The evidence at trial confirms that these differences are due to the Search Function Policy. Because cooperating brokers have an incentive to look at all listings that might meet their clients' needs, Mr. Murray (the industry expert) explained that he could attribute this decline to the Search Function Policy. (CCPF 921-22). Limited service brokers also uniformy testified about numerous complaints from customers that their listings "didn't show up" on the Realcomp MLS. (CCPF 923-26). As one limited service customer wrote, I've called 2 separate real estate agents just to see if they could locate my listing on the MLS. In both of their searches my listing did not come up. The only way -30­ it was found was by entering the MLS number. Can you tell me why this is happening??? What good is it to have it on the MLS ifit doesn't come up in a search?? (RX 67-005; CCPF 933). Anotherlimited service customer complained that her Exclusive Agency listing "did not appear in a regular MLS search for a house with my propert's characteristics." (RX 40-002; CCPF 988). Moreover, limited service brokers testified that they received calls (several a week) from other Realcomp brokers who could not find the Exclusive Agency listings through a search on the Realcomp MLS. (CCPF 964-65, 986-90, 1018-19, 1046­ 48). These brokers do not receive similar complaints or calls regarding their Exclusive Agency listings in other MLSs. 15 This evidence is not surprising. Study after study confirms that defaults matter. (CCPF 937-40). Whether it is a pre-checked box (as in the Search Function Policy) or a pre-chosen answer, studies show that defaults make a substantial difference in people's actions. 2. Realcomp's Policies Affect Discount Brokers' Abilty to Compete Limited service brokers uniformly testified that the Realcomp Policies impacted their ability to compete in the Realcomp service area. (CCPF 861-1068). The evidence shows that Realcomp Policies: · Caused market exit. Y ourlgloo, a limited service brokerage that offers services in several states, ceased doing business in Michigan due to Realcomp's policies. the top two or (CCPF 954-71). Even though the Michigan market was one of three markets, Y ourlgloo exited wholly due to the problems caused by Realcomp's Policies. (CCPF 968; CX 422 (Aronson, Dep. at 119 (attibuting YourIgloo's exit "(o)ne hundred percent" to Realcomp's Policies))). 15 The Realcomp Board of Governors even received a request to change the default because brokers did not realize that default searches only resulted in Exclusive Right to Sell (CCPF 927). At least one Realcomp Governor voted to change the default because of listings. the unfairness of the system. (CCPF 928). -31­ · Prevented entry. BuyS elf Realty, another limited service brokerage that offers services in several states, was deterred from entering the Realcomp market area because ofRealcomp's Policies. (CCPF 972-1006). As Albert Hepp testified at tral, after studying Realcomp's rules and experiencing complaint after complaint from consumers who his business had referred to limited service brokers in the Realcomp area, he determined not to enter that area directly because "( w)e can't do it with the existing rules as they are." (Hepp, Tr. 653-54; CCPF 1006). · Hampered competition from the remaining discount brokers. The Policies severely degrade the quality of the services that discount brokers can offer using Exclusive Agency listings and forced these brokers to develop more expensive packages using Exclusive Right to Sell listings. (CCPF 1007-68). Due to Realcomp's Policies, discount brokers cannot deliver the wide market exposure consumers demand, at least not with Exclusive Agency listings that offer substantial savings.16 (CCPF 868-74, 894-98). Instead, they are forced by Realcomp's Policies to explain the limitations placed on Exclusive Agency listings. (CCPF 1042-43, 1051-54, 1059-63, 1065-68). For instance, although Craig Mincy of MichiganListing wants to "show the general public they don't necessarily have to pay 6 percent to sell their home" by using an Exclusive Agency listing, he has to explain to potential customers that there is "no way of gettng their listing. . . onto MovelnMichigan.com or. . . the IDX system." (Mincy, Tr. 377-78,422-23; CCPF 1013, 1025). Further, most of his potential customers ask whether their listings wil be "as accessible as everybody else's" on the Realcomp MLS, so Mr. Mincy must explain that "there is an extra step that agents have to take, and if they don't take it, they're not going to see your listing, and so that 16 Discount brokers are able to get these listings to Realtor.com, but because ofthe additional costs due to double listing, they often charge more for this service. (CCPF 1021, 1033, 1038-39; CX 435 (Greater Michigan Realty charges more for Exclusive Agency package with Realtor.com); CX 439 (MichiganListing.com charges an extra $100 for access to Realtor.com)). -32­ is a problem." (Mincy, Tr. 423; CCPF 1025-26). Because ofthese issues, Mr. Mincy has "lost a substantial amount of business." (Mincy, Tr. 425; CCPF 1027-28). Jeffry Kermath of Amerisell Realty informs customers of these problems through another means. He posts a detailed description of the issues on his website, which is a must read for consumers in "Wayne, Oakland, Macomb, or Livingston Counties." (CCPF 1059; RX 12-007). His site explains that because of Realcomp' s Policies: 1. Exclusive Agency listings are not included in the default search on the Realcomp MLS, thus "you wil not be found some of the time when Realtors do searches." 2. Realcomp wil not allow these listings to go to MovelnMichigan. "This is huge! This is Realcomps public MLS site where thousands of people search!" 3. Realcon;p wil not allow these listings on the Realcomp IDX. "Many people these days surf Realtor websites and unless you're an erts listing, you wil not be found. " 4. Realcomp wil not allow these listings to go to Realtor.com. (RX 12-007; CCPF 1063-68). To mitigate these problems, discount brokers have developed packages based on Exclusive Right to Sell contracts, which must include the five minimum services required by Realcomp's Policies. (CCPF 1012, 1034). Greater Michigan Realty, for instance, offers three packages based on Exclusive Agency listings and two. Exclusive Right to Sell listings. (CX 435; CCPF 1032). The Exclusive Agency packages allow the seller to avoid paying the offer of compensation ifno cooperating broker is involved. (CCPF 1229). Under the Exclusive Right to Sell packages, however, sellers do not retain the right to sell the propert on their own, which means that they must pay the offer of compensation whether or not a cooperating broker is involved. (CCPF 1229). Greater Michigan also provides the five minimum services mandated -33­ by Realcomp's Policies. (CCPF 1034). These packages are thus more expensive than the Exclusive Agency packages; Greater Michigan charges a higher flat fèe as well as a 0.5% listing commission. (CX 435; see also CCPF 1012, 1029 (MichiganListing.com Exclusive Right to Sell packages more expensive); CCPF 1053 (Exclusive Right to Sell packages more expensive)). Limited service brokers testified at trial that the more expensive Exclusive Right to Sell listings, which obtain full exposure through the Realcomp MLS, enjoyed better success than their Exclusive Agency listings on the Realcomp MLS. (CCPF 1037-41, 1054-68). In contrast, their Exclusive Agency listings in other MLSs that do not have discriminatory rules receive as much activity as Exclusive Right to Sell listings and are more successful than Exclusive Agency listing in the Realcomp MLS. (CCPF 1037, 1044-45, 1118-22). In short, due to Realcomp's Policies, limited service brokers can only offer handicapped Exclusive Agency listings or more expensive Exclusive Right to Sell listings bundled with "full service." They cannot offer the package consumers want - an Exclusive Agency listing with full exposure - and this impacts their ability to compete effectively. 3. Realcomp's Policies Reduced the Use of Discount Brokerage Services on a Market-Wide Basis These are not isolated effects. Data from the Realcomp MLS and a number of other MLSs demonstrate that Realcomp's Policies have a market-wide impact. Even though there is some evidence that particular discount brokers have managed to increase their business, the data show that Rea1comp's Policies reduce the overall use of Exclusive Agency listings (and therefore discount brokerage services) in the Realcomp MLS. (CCPF 1069-1116). -34­ Realcomp Time-Series Analysis A time-series analysis of Realcomp's listing data shows that the share of Exclusive Agency listings in the Realcomp MLS dropped after Realcomp made identifying listing type mandatory in late 2003 and early 2004. (CCPF 1080-84). Since the Realcomp Website Policy was adopted in 2001, this analysis likely measures the decline in Exclusive Agency listings from an already reduced level, and therefore understates the effect of Realcomp's Policies. (CCPF 1083). Nonetheless, the available data show that the share of Exclusive Agency listings dropped 50%, from approximately 1.7% in early 2004 to less than 0.8% in late 2006. (CCPF 1081). As Realcomp's economist admitted, this decrease cannot be attributed to conditions in the real estate market. (Eisenstadt, Tr. 1621-22; CCPF 1084). Benchmark Analyses Benchmark analyses comparing data from the Realcomp MLS with data from several MLSs with and without similar restrictions also show that Realcomp's Policies reduce the use of Exclusive Agency listings. (CCPF 1085-97). Dr. Wiliams ranked every Metropolitan Statistical Area in the countr on whether they were "similar" to the Realcomp area based on certain economic and demographic variables that may theoretically be related to the use of Exclusive Agency listings. (CCPF 1086). He then obtained five years' worth oflisting data from each of the top six MLSs in that rankng that did not have any policies similar to Realcomp's Policies. (CCPF 1086-90). In addition, Dr. Wiliams collected data from three MLSs that had website policies similar to Realcomp's. (CCPF 1089). -35­ The results unambiguously show that Realcomp and the MLSs with similar restrctions had a signficantly smaller share of Exclusive Agency listings in each of the five years in the sample. (CCPF 1097). Comparison of MLSs With and Without Acces Restrictions Percent of NON-ERTS Listings 9.0% 8.0% 7.0% 6.0% .: 5.0% r ~ il l4.0% 3.0% 2.0% 1.0% 0.0% ç ~ ï§ ï~ q ~ i§ i~ q ~ i~ i~ ~:s î~ i~ 9 ~ î§ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ i ~ ~ ~ ~ ~ i ~ ~ ~ ~ ~ i ~ g~ -R.c:omp .. Mls with Restrctions -- MIs without Restrctons a x '" ~ Sour: MLS Listig Data frm various MLSs. (CX 524). Statistical Analyses Statistical analyses confirm that Realcomp's Policies are associated with a substantial reduction in the use of Exclusive Agency listings in the Realcomp MLS. (CCPF 1098-1104). To ensure that the Realcomp Policies (and the similar policies in the other MLSs) caused the differences in use of Exclusive Agency listings, Dr. Wiliams conducted a number of statistical analyses to control for such factors as housing characteristics, changes over time, demographic factors, the state of the housing market, and economic factors. (CCPF 1098-1100). Dr. Wiliams also reran his own statistical analyses adding the economic and demogrphic variables that -36­ Respondent's economist believed were significant. (CCPF 1101). Each of these analyses resulted in substantially similar results. (CCPF 1102-03). The analyses showed that Realcomp's restrctions were associated with a reduction in Exclusive Agency listings of 5.5 to 5.8 percentage points. (CCPF 1102). These statistical analyses show that but for the Realcomp Policies the expected share of Exclusive Agency listings in the Realcomp MLS would be approximately 6 to 7%. (CCPF 1104). Realcomp's Economist's Analyses Realcomp's own economist conducted both time series and benchmark analyses that are consistent with Dr. Wiliams analyses. (CCPF 1105-13). Dr. Eisenstadt's time-series analysis for the Boulder MLS shows that a website policy there decreased the percentage of Exclusive Agency listings over 50%, from an average of2.03% prior to the restrction to an average of 0.98% after the restriction. (CCPF 1106-07). Dr. Eisenstadt's benchmark analysis using the An Arbor MLS also shows an effect by the Realcomp Policies. (CCPF 1108). Data for the entire An Arbor MLS show a share of Exclusive Agency listings at 4.2% compared to 0.74% for Realcomp. (CCPF 1108). Dr. Eisenstadt also examined only Washtenaw county listings in the Ann Arbor MLS, but the data stil show a higher share of Exclusive Agency listings in the An Arbor MLS (1.6% versus 0.74%). (CCPF 1109-11). Furher, on cross-examination, Dr. Eisenstadt recalculated the Ann Arbor MLS data to exclude only listings from the Realcomp area (Oakland, Wayne, Livingston, and Macomb), resulting in a 3.59% share in the An Arbor MLS. (CCPF 1112-13). * * * -37­ In sum, all of this empirical evidence points to the same conclusion - the Realcomp Exclusive Agency listings in the Realcomp MLS. Policies substantially reduce the share of H. Realcomp's Agreement to Change Its Search Function Policy but Refusal to Change its Website Policy Durng tral, the parties negotiated a stipulation for entr of an order regarding Realcomp's Search Function Policy. The stipulated order would preclude Realcomp from discriminating against Exclusive Agency listings in its search function. It would also prohibit Realcomp from requiring minimum services for Exclusive Right to Se1l1istings. The stipulation does not affect Realcomp' s Website Policy, and Realcomp does not admit liability. The Realcomp Board of Governors has made it clear that Realcomp wil continue to exclude Exclusive Agency listings from Realcomp's feed of listing information to public websites and the IDX websites. At the same time it voted to change the Search Function Policy, the Realcomp Board rejected a motion that would have eliminated the Website Policy. (CCPF 845, 859-60). That motion proposed that Realcomp comply with a recently passed NAR rule that requires NAR associated MLSs to include all listings, regardless oflisting tye, in any feed to public websites and in the IDX. (CCPF 839-41). Compliance with this rule is mandatory, which means that NAR has determined that the rule is necessary for the proper operation of an MLS. (CCPF 841, 844). Realcomp had urged NAR not to pass this mandatory rule, but NAR rejected Realcomp's arguments. (CCPF 845-58). The stipulated order, therefore, addresses only the remedy regarding the Search Function Policy. It does not resolve the overall issue ofliability for Realcomp's conduct. See, e.g., United States v. Uniroyal, Inc., 300 F. Supp. 84, 100 (S.D.N.Y. 1969) (finding that determining liability -38­ for a Sherman Act violation is a "somewhat different" issue than that used to determine whether the governent is entitled to injunctive relief); see also United States v. Parke, Davis & Co., 365 u.s. 125, 126 (1961). The remaining issues before the Cour are therefore (1) Realcomp's overall liability for its conduct and (2) the appropriate additional remedial measures necessary to restore competition. III. LEGAL ANALYSIS Realcomp's Policies violate Section 5 of the FTC Act. The record shows that (1) the Policies are the result of a combination of competitors (Realcomp's members); (2) the nature of the Policies is to restrain competition from discount brokers and to eliminate a product that consumers desire; (3) Realcomp has the power to restrain competition among brokers; (4) the Policies have had anticompetitive effects; and (5) the Policies have no procompetitive justification. The Court should therefore enjoin Realcomp from continuing to restrain competition. A. The Antitrust Laws Seek to Protect Competition to Ensure Lower Prices, Higher Quality, and Greater Choices for Consumers The antitrst laws prohibit combinations of competitors that restrain trade because competition is good for consumers. As the Supreme Cour has explained, the Sherman Act proscribes unreasonable restraints on competition to ensure lower priced, higher quality products for consumers: The Sherman Act was designed to be a comprehensive charter of economic libert aimed at preserving free and unfettered competition as the rule of trade. It rests on the premise that the unrestrained interaction of competitive forces wil yield the best allocation of our economic resources, the lowest prices, the highest quality and the greatest material progress . . . -39­ Northern Pac. Ry. Co. v. Us., 356 U.S. 1,4(1958). The cours therefore routinely condemn agreements among competitors that result in higher prices, lower output, or lower quality. The antitrust laws, however, are especially concerned with competitor combinations that unreasonably limit consumer choice. As the Supreme Court put it, "since Congress designed the Sherman Act as a 'consumer welfare prescription, '" the "most significant" harm from competitor agreements to restrain competition is a market that is "unesponsive to consumer preference." NCAA v. Bd. of Regents of the Univ. Okla., 468 U.S. 85, 106-07 (1984) (internal quotes and citation omitted). Thus, in FTC v. Indiana Federation of Dentists, 476 U.S. 447 (1986), the Supreme Court condemned an agreement among competitors to "withhold from their customers a particular service that they desire" because "(a)bsent some countervailing procompetitive virte. . . such an agreementlimiting consumer choice by impeding the 'ordinary give and take of the market place,' cannot be sustained." Id. at 459. Combinations of competitors that limit consumer choice in this way uneasonably restrain trade. See Glen Holly Entm 't Inc. v. Tektronix Inc., 352 F.3d 367,374,378 (9th Cir. 2003) (holding that the antitrst laws protect consumers from competitor agreements that limit consumer choice by eliminating a competing product); Welchlin v. Tenet Healthcare Corp., 366 F. Supp. 2d 338,353 (D.S.C. 2005) (holding that "depriving consumers of fair choice among providers of medical services" is competitive harm). B. The Elements of Complaint Counsel's Claims and the Burden of Proof The Complaint alleges that Realcomp's Policies are precisely the tye of conduct the antitrst laws condemn - a combination or conspiracy of competing brokers that unreasonably restrains trade. (CCPF 12; (Complaint at ~~ 24,27)). As the parties have stipulated, the elements of this claim are: (1) the existence of a contract, combination, or conspiracy among two -40­ or more separate entities, that (2) unreasonably restrains trade, and (3) affects interstate commerce or foreign commerce. 17 (JX 1-09 (citing Law v. NCAA, 134 F.3d 1010,1016 (10th Cir. 1998) (identifying elements of a violation of Section 1 of the Sherman Act); Fashion the FTC Act Originators' Guild, Inc. v. FTC, 312 U.S. 457, 463-64 (1941) (Section 5 of violations may be based on conduct that violates the Sherman Act)). Realcomp has admitted two of these elements. Realcomp admits that the Policies affect interstate commerce. (CX 32-003 (Answer at ir 6)). Realcomp has also stipulated that it "is a combination of its members with respect to the policies at issue." (JX 1-10; CCPF 12); see also us. v. Realty Multi-List, 629 F.2d 1351, 1361 n.20 (5th Cir. 1980) (holding that members of an MLS engaged in the "concerted action necessar to establish a Section 1 violation" by adopting and applying MLS rules); In re North Texas Specialty Physicians, Dkt No. 9321,2005 FTC LEXIS 173, at *36 (F.T.C. Nov. 29, 2005) ("The Commission has also held that when an organization is controlled by a group of competitors, the organization is viewed as a combination of its members, and their concerted actions wil violate the antitrust laws if an unreasonable restraint of trade."). Thus, Complaint Counsel need only prove that Realcomp's Policies unreasonably restrained trade. Agreements unreasonably restrain trade when they have (or are likely to have) a substantial anticompetitive effect in the relevant market, such as by increasing prices, reducing output, reducing quality, or reducing consumer choice. See, e.g., Standard Oil Co, v. United States, 283 U.S. 163, 179 (1931); Hahn v. Oregon Physicians' Serv., 868 F.2d 1022, 1026 (9th 17 Realcomp also admits that the Commission has jurisdiction over this matter. (CX 32-003 (Answer at ir 6); ix 1 -09). -41­ Cir. 1988). To determine whether a restraint likely harms competition, courts consider one or more of three factors: the nature of the restraint, market power, and evidence of actual effects. In many cases, the likely har to competition is apparent from the nature of the restraint. F or instance, some tyes of restraints, such as naked price-fixing agreements, are per se ilegaL. Similarly, restraints that bear a "close family resemblance" to "another practice that already stands convicted in the court of consumer welfare" are presumed to harm competition. Polygram Holding, Inc. v. FTC, 416 F:3d 29,37 (D.C. Cir. 2005). Thus, if "based upon economic learning trade likely impairs reason analysis. Id. at 36. The courts and the experience of the market, it is obvious that a restraint of competition," the courts wil apply an abbreviated rule of wil presume harm to competition from the natue of the restraint, without any need for proof of market power or actual effects.18 Id. Even where the anticompetitive nature of a restraint is less obvious, the cours need not engage in a full rule of reason or "plenar market examination." California Dental Assoc. v. FTC, 526 U.S. 756, 779 (1999) (the need for "a more extended examination of the possible factual underpinnngs. . . is not, of course, necessarily to call for the fullest market analysis"). In examining agreements among competitors, the essential inquiry is "whether or not the challenged restraint enhances competition." Id. at 779-80. The cour need only conduct a sufficient analysis to arrive at a "confident conclusion about the principal tendency of a restriction." Id. at 781. 18 For instance, the Commssion condemned under an abbreviated rule of reason analysis ajoint venture's moratorium on discounting and advertising for products outside of the venture, In re Polygram Holding, Inc., 2003 FTC LEXIS 120 (Iul. 24, 2003), and a licensing board's ban on advertising discounts by optometrists, Massachusetts Bd. of Registration in Optometry, 110 F.T.C. 549, 607 (1988). -42­ Courts have therefore repeatedly held that evidence regarding the nature of the restraint combined with evidence of market power is tyically suffcient to presume anti competitive Stationery & effects. See, e.g., Northwest Wholesale Stationers, Inc. v. Pacifc Printing Co., 472 u.s. 284, 296-98 (1985) (to establish a presumption of anticompetitive effects when conduct not "virtally always likely to have an anticompetitive effect" requires "some showing" of "market power or unque access to a business element necessary for effective competition"). The natue of the restraint and market power may therefore establish presumed anticompetitive effects, even in the absence of proof of actual anticompetitive effects, such as higher prices. See, e.g., Indiana Fedn of Dentists, 476 U.S. at 461-62 (1986) (holding that under abbreviated rule of reason analysis, restraint could be condemned "even absent proof that it resulted in higher prices or, as here, the purchase of higher priced services, than would occur in its absence"); United States v. Brown Univ., 5 F.3d 658,668 (3d Cir. 1993) (explaining that proof of market power is sufficient to create presumption of anticompetitive effects without proof of actual anticompetitive effects). If the nature of the restraint and evidence of market power show likely harm to competition, the respondent bears the burden of showing that the challenged policies have a "plausible" and "cognizable" procompetitive justification. See In re Polygram Holding, Inc., 2003 FTC LEXIS 120 at *62 (Jul. 24, 2003). A justification is cognizable ifis compatible with the goal of the antitrust laws to fuher competition. Id. at *65. It is plausible if it wil "plausibly create or improve competition" (such as by increasing output or improving product quality, service, or innovation) and "cannot be rejected without extensive factual inquiry."19 Id. 19 "The defendant, however, must do more than merely assert that its purported justification benefits consumers. Although the defendant need not produce detailed evidence at this stage, it must articulate the specific link between the challenged restraint and the purprted justification to merit a -43­ Moreover, the challenged policies must be reasonably necessary (and no broader than necessary) to achieve the procompetitive effects. See, e.g., NCAA, 468 U.S. at 1l4-15(explaining that restraint must be least restrictive means reasonably available to achieve alleged effciencies); Polygram, 2003 FTC LEXIS 120, at *61-62; Law, 134 F.3d at 1019; In re Brunswick Corp., 94 F.T.C. 1174, 1275 (1979). Absent such a showing, the restraints are condemned. The courts have long applied this abbreviated rule of reason analysis to restrictive MLS rules. In United States v. Realty Multi-List, Inc., 629 F.2d 1351 (5th Cir. 1980), the Departent of Justice challenged the membership criteria of an MLS, which required that members have a "favorable credit report and business reputation" and maintain an offce "kep! open during customary business hours" in a particular county. Id. at 1358. To evaluate these restraints, the Fifth Circuit set forth an abbreviated rule of reason analysis that "allows the cours to reach and void on its face any significantly restrictive rule of a combination or trade association with significant market power, which lacks competitive justification or whose reach clearly exceeds the combination's legitimate needs." Id. at 1370. Under this analysis, once market power is shown, the burden of proof is on the MLS to justify its rule. See Thompson v. Metropolitan Multi-List, Inc., 934 F.2d 1566, 1581 (11 th Cir. 1991). The cour may then "determine the reasonableness of an association's rule on its face by gauging its justification in terms of the competitive needs of the association and by examining the rule itself to determine if it is drawn in such a manner as to further that need without unnecessarily trampling competitive opportities." Realty Mult-List, 629 F.2d at 1372. If the more searching inquiry into whether the restraint may advance procompetitive goals, even though it facially appears of the type likely to suppress competition." Polygram, 2003 FTC LEXIS 120, at *65. -44­ MLS rules are not "reasonably necessary" to the "competitive needs of the association" and "narrowly tailored to that end," the rule "may be condemned on its face, without proof of past effect." Id. at 1375. Here, the evidence leaves no doubt that Realcomp's Policies do not enhance competition but restrain it. The Policies by nature restrct competition, and Realcomp has market power. Realcomp's Policies have no plausible or cognizable efficiency justifications, and should therefore be condemned under an abbreviated rule of reason analysis. But the evidence also shows (should Your Honor reach the issue) that Realcomp has uneasonably restrained trade under an even fuller market enquiry; the record shows that Realcomp's Policies have actual adverse effects on competition. C. Realcomp Unreasonably Restrains Competition in the Real Estate Brokerage Services Market and Has Violated Section 5 of the FTC Act 1. Realcomp's Policies are the Result of an Agreement Among Competing Real Estate Brokers Realcomp's admission that it "is a combination of its members with respect to the policies at issue,"20 (JX 1-10), has important implications for the competitive analysis of the Policies. The Supreme Court has cautioned that combinations of competitors are "inherently fraught with anticompetitive risk." Copperweld Corp. v. Independence Tube Corp., 467 U.S. 20 This conclusion was inevitable. Realcomp is owned by seven associations of competing real estate brokers. These associations of competitors appoint the members of Realcomp' s Board of Governors. The Board, which is comprised of competing real estate brokers, sets Realcomp' s rues and policies. And Realcomp's members are competitors in the market for real estate brokerage services. (CCPF 205,257,267,279); see, e.g., Alvord-Polk, Inc. v. F. Schumacher & Co., 37 F.3d 996,1007 (3d Cir. 1994) (holding that association action taken on behalf of its competing members, such as when a board of directors or a committee adopts a rule or policy, is considered to be the concerted action of the competing members). -45­ 752, 768-69 (1984). Thus, associations with members who are competitors "have traditionally been the objects of antitrt scrutiny" because the members "often have economic incentives to restrain competition."21 Alled Tube & Conduit Corp. v. Indian Head, Inc., 486 U.S. 492,500 (1988). As Realcomp' s own economist explained, "Competitors naturally wil try to restrict each other's output, either by forming a collusive combination or by driving one another out. Consequently, antitrust is rightly suspicious of any horizontal 'restraint of trade.'" (Eisenstadt, Tr. 1523-24). These principles apply with special force in the case of an MLS. Because an MLS such as Realcomp "may create significant competitive advantages for both its members and for the general public, there exists the potential for significant competitive harms when the group, having assumed significant power in the market, also assumes the power to exclude other competitors from access to its pooled resources." Realty Multi-List, 629 F.2d at 1370. This tye of collaboration is typically subject to a "higher degree of scrutiny" because it is well recognized that sellers often "face(J a severe disadvantage if denied access." 13 PHILLIP E. AREEDA & HERBERT HOVENKAMP, ANTITRUST LAW §§ 2220b2, 2215e (2006) ("ANTITRUST LAW"). As explained by the. Supreme Court, competitor collaborations with these tyes of effects cannot use the advantages achieved by collective means to suppress competition. Associated Press v. United States, 326 U.S. 1, 18-19 (1945). 21 As the leading antitrust treatise has explained, for instance, members of an MLS have wil increase the output of 13 PHILLIP E. AREEDA & incentives to use the MLS to disadvantage a discount broker because, even though the discount broker the MLS overall, the broker wil take market share away from oth",r members. HERBERT HOVENKAMP, ANTITRUST LAW § 2220c (2006). -46­ 2. The Nature of Realcomp's Policies is to Restrict Competition The record shows that this higher level of scrutiny is deserved; Realcomp' s Policies restrict competition in two ways. First, the Realcomp Policies tend to exclude competition from discount brokers by disadvantaging the use of their primary competitive tool- the Exclusive Agency listing agreement. The Policies therefore deny consumers the benefits of this competition. Second, the Policies limit competition among Realcomp members by eliminating their ability to offer a particular package of services - Exclusive Agency listings with full exposure through the Realcomp MLS. The Policies thus deny consumers a product that they desire. a. Realcomp's Policies Harm Competition and Consumers By Restricting Competition from Discount Brokers The courts have long recognized the anti competitive potential of MLS rules that deny MLS services to some brokers. See, e.g., Realty Mult-List, 629 F.2d at 1370-71; Thompson, 934 F.2d at 1580. While these cases generally deal with the denial of membership, the principles developed in these cases apply here.22 And the evidence shows that the competitive hars . identified by the cours in these cases are the same as those here. In Realty Multi-List, the Fift Circuit explained that rules that uneasonably deny multiple listing serices to some brokers harm those brokers' ability to compete because they cannot "gain wide exposure of their listings." Realty Multi-List, 629 F.2d at 1370-71. Other courts have 22 As a leading treatise points out, "product exclusion" - "when a venture disapproves a particular product, or decides not to permit the product to be produced within the venture" - can be as anti competitive as "member exclusion." 13 ANTITRUST LA W § 2220b3 (2006). Product exclusion is anticompetitive when its "effect is to keep off the market a product or process that consumers would prefer." Id. Here, the competitor collaboration decided not to "produce" Exclusive Agency listings with full exposure through the Realcomp MLS. -47­ echoed these conclusions. See, e.g., Marin County Bd. of Realtors v. Palsson, 16 CaL. 3d 920, 935-36 (1976) (rule denying multiple listing services to part-time brokers "seriously hampers the competitive effectiveness of nonmember licensed brokers"). For instance, as one court explained, because consumers "naturally desire(J the widest market exposure possible," they are unlikely to use a broker denied MLS services. Oates v. Eastern Bergen County Multiple Listing Service, 273 A.2d 795,800 (N.J. Super. Ch. 1971). These tyes of rules also harm consumers. According to the Fifth Circuit: Buyers and sellers are also harmed by unjustified exclusions. Even though member brokers still compete with each other to procure listings and to sell any listing in the pool, the public is denied the incentive to competition that new entry may bring. A new entrant into the market might, for example, be more aggressive and wiling to accept a lower commission rate. Exclusion of such a broker would tend to reduce the amount of price competition in the market. Realty Multi-List, 629 F2d at 1371 (citations omitted). In addition, the court recognized that under these MLS rules, consumer choice would be limited in another way: A person wishing to sell or buy a home may believe that a particular nonmember is more competent than available members. But if the consumer wishes to have ready access to a large market in a short .period of time, he may be forced to deal with a less desirable member broker or salesman. Id. As the California Supreme Cour put it, by makng it diffcult for brokers denied services to compete, such rules "tend to limit entry into a competitive field (and) . . . . (c)onsumer choice is thereby narowed." Palsson, 16 CaL. 3d at 937. The Eleventh Circuit succinctly stated these competitive harms: "First, the excluded broker's listings wil not be distributed as widely as possible, resulting in inefficient sales prices. Second, the exclusion reduces the competition -48­ among brokers and could result in less competition for brokerage fees." Thompson, 934 F.2d at 1580. Although Realcomp allows limited service brokers to join Realcomp and post Exclusive Agency listings on the MLS, Realcomp's Policies cause competitive harm similar to the denial of membership. Courts have long recognized that the denial of some services of a competitor collaboration can lead to the same competitive harm as a denial of all services. In Northwest Wholesale Stationers, for instance, the rules of the competitor collaboration (a buying cooperative) did not fully exclude competitors. 472 U.S. at 286,297 n.6 (all retailers could purchase through the collaboration at the same price, but only members received a year-end rebate). Yet the Cour recognized that a rule providing only discriminatory access to the collaboration's services - a concerted refusal to deal on substantially equal terms - would violate the antitrust laws if it nonetheless restricted competition. Id. at 297 n.6 (citing Joseph Brodley, Joint Ventures and Antitrust Policy, 95 HARv. L. REV. 1521, 1532 (1982) ("Even if the joint venture does deal with outside firms, it may place them at a severe competitive disadvantage by treating them less favorably than it treats the (participants in the joint ventue)")). The issue is not whether the competitor is fully excluded but whether competition has been restrained: The relevant question is not whether the plaintiff has or has not been "excluded," or whether he is or is not a member after the challenged rule has been applied. The issue is whether the challenged rule facilitates a restraint of trade, and this requires a showing that the plaintiff is suffciently disadvantaged to permit the defendants to effect a marketwide output reduction. This can certinly happen if the plaintiff is excluded from the market altogether, but it can also happen if the plaintiff faces significantly higher costs or a significantly less attactive product or distribution offering than previously. -49­ 13 ANTITRUST LAW § 2214e; see also id. at § 2220b3 (collaboration's refusal to produce product used by competitor may have similar impact as excluding competitor as member); Kreuzer v. American Acad. of Periodontology, 735 F.2d 1479, 1482-84, 1494 (D.C. Cir. 1984) (identifyng harm from refusal to deal with competitor on substantially equal terms). The evidence shows that these principles are applicable here. The ability of discount brokers to compete for listings is restrained because, unlike full-service brokers' Exclusive Right to Sell listings, Exclusive Agency listings do not receive "the widest market exposure possible." Oates, 273 A.2d at 800; see also Thompson, 934 F.2d at 1580 ("broker's listings wil not be distributed as widely as possible, resulting in inefficient sales prices"). As the record reflects: · Exposure is "critical" to sellng a home. (CCPF 454-59). · Sellers recognize the importance of exposure and demand it. (CCPF 460-62). · Sellers therefore expect their home to be listed on the local MLS. (CCPF 478-81). The Realcomp MLS in particular exposes its members' listings to approximately 14,000 other members - more than any other MLS in Michigan. (CCPF 279, 282, 504). · The Search Function Policy limited this key exposure on the Realcomp MLS. (CCPF908-36).23 · Marketing listings on the Internet is also criticaL. (CCPF 536, 673-76). Eighty red listings are marketed on the Internet. (CCPF 573). Consumers demand it. (CCPF 592, 654, 664, 870, 943). This is no doubt because 80% of buyers are searching the Internet as part of their home search and 24% of buyers purchase a home they first found on the Internet. (CCPF 541-42, 553-54). five percent of broke 23 As courts have recognized, search defaults can have negative competitive effects even when they are easy to overrde. See, e.g., United Air Lines, Inc. v. Civil Aeronautics Bd., 766 F.2d 1107, 1110, 1113 (7th Cir. 1985) (Posner, J.) (recognizing competitive effects of search default to list certain air fares first in travel agent's computer as diverting business to those airlines even though "(I)oyal and skilful travel agents no doubt correct for the tilt" by simply pushing a button, "but not all travel agents are either")); (cf CCPF 937-40). -50­ · Studies have repeatedly found that the best categories of web sites for marketing properties are those covered by the Approved Websites: MLS public websites (MoveInichigan); Realtor.com; and broker and agent web sites (Realcomp member IDX websites). (CCPF 594-600). While brokers may post their listings on other websites, "the ones ( a broker) really ha( s) to be on to compete effectively are the four major sites where 40 to 50 percent of buyers are going." (Murray, Tr. 238 (referring to MLS websites, Realtor.com and IDX websites); CCPF 673). · The Website Policy limits this important exposure by excluding Exclusive Agency listings from Realcomp' s feed of listing information to these key categories of web sites. (CCPF 865-74). · The only way for discount brokers to overcome the significant competitive disadvantage of Sell the Website Policy is to offer more expensive Exclusive Right to listing agreements with the five minimum services that Realcomp requires. (CCPF 667,875,330-33). Put simply, Realcomp's Policies harm consumers because they "reduce() the competition among brokers and could result in less competition for brokerage fees," Thompson, 934 F.2d at 1580, and limit consumer choice, Realty Multi-List, 629 F.2d at 1371; Palsson, 16 CaL. 3d at 937. The evidence therefore shows that the Realcomp Policies cause the same tye of competitive harm as do rules excluding brokers from membership in the MLS. Moreover, even in cases not involving exclusion from an MLS, courts have condemned MLS rules that prevent brokers from using effective means of gaining exposure for their listings. See Cantor v. Multiple Listing Servo of Dutchess Cty, 568 F. Supp. 424, 427 n.6, 430 (S.D.N.Y. 1983) (holding that effective yard signs was unreasonable restraint of MLS rule preventing use of trade). b. Realcomp's Policies Harm Competition and Consumers By Eliminating a Product Desired By Consumers In addition to effectively excluding competition from discount brokers, Realcomp's Policies also limit competition among Realcomp members by a collective agreement not to offer consumers Exclusive Agency listings with full exposure through the Realcomp MLS. In other -51­ words, through the Policies, Realcomp members have agreed to limit the package of services that they wil offer and with which they wil compete for customers. The Supreme Cour has made clear that this type of agreement - an agreement among competitors to "withhold from their customers a particular service that they desire" - requires only "a rudimentary understanding of economics" to conclude that the agreement "would have an anti competitive effect on customers and markets." CaL. Dental Assoc., 526 U.S. at 770. In Indiana Federation of Dentists, for instance, insurance companies sought to offer consumers "alternative benefits" dental insurance, which required a review of requests for dental treatment before any procedure. Id. at 449. To implement these plans, "dentists' customers (that is, the patients and their insurers) sought a particular service: cooperation with the insurers' pretreatment review through the forwarding ofx rays in conjunction with claim forms." Id. at 457. Local dentists, however, collectively agreed to refuse to provide this service, denying "the information the customers requested in the form that they requested it." Id. Applying a "quick look" rule of reason analysis, the Court concluded that such agreements to "withhold from their customers a particular servce that they desire" unreasonably limit consumer choice: A refusal to compete with respect to the package of services offered to customers, no less than a refual to compete with respect to the price term of an agreement, impairs the ability of the market to advance social welfare.,. . . Absent some countervailing procompetitive virte - such as, for example, the creation of effciencies in the operation of a market or the provision of goods and services . . . such an agreement limiting consumer choice by impeding the 'ordinary give and take of the market place,'. . . cannot be sustained under the Rule of Reason. Id. at 459; see also United States v. VISA U.S.A., Inc., 344 F.3d 229,242 (2d Cir. 2003) Goint ventue rules prohibiting members from competing "with the others in a manner which the -52­ consortium considers harmful to its combined interests" was exemplar of anti competitive bahavior). Without evidence of any procompetitive justification, the Court condemned the agreement, even absent proof of market power. Id. at 459-60; see also Cal. Dental Assoc., 526 u.s. at 770 (explaining that Cour applied "quick look" analysis, which does not require proof of market power). The Court also held that the natue of the restraint obviated the need for proof of actual anti competitive effects, such as higher prices, because the agreement "is likely enough to disrupt the proper fuctioning of the price-setting mechanism of the market that it may be condemned even absent proof that it resulted in higher prices or, as here, the purchase of higher priced services, than would occur in its absence." Id. at 461-62. Moreover, "even if the desired information were in fact completely useless," the Court reasoned that competitors were "not entitled to pre-empt the working of the market by deciding for itself that its customers do not need that which they demand." Id. at 462. Realcomp's Policies similarly restrict competition. The evidence shows that consumers want Exclusive Agency listings with full exposure through the MLS, but this package of services is simply not available to them because a combination of competing brokers agreed to no longer offer it: · Realcomp members post their listings on the Approved Websites for the additional exposure, advertse their Internet marketing skills to potential clients, and have never had a customer ask to not be placed on the Internet websites. (CCPF 581-87, 654, 1172-73). · Customers expect that their homes wil be displayed on the Approved Websites. (CCPF 870, 1164-73). In fact, exposure on IDX websites is becoming "more and more" important to customers. (G. Moody, Tr. 827, 831; CCPF 654). -53­ · Discount brokers received numerous complaints from customers using Exclusive Agency listings who wanted to be on the Approved Websites, but were not because of Realcomp's Website Policy. (CCPF 962-65, 991-92, 1042). · Because of the Website Policy, many consumers using Exclusive Agency listings paid more for their listing to be exposed on Realtor.com. (CCPF 1157-58; D. Moody, Tr. 493-94; CX 435; see also Mincy, Tr. 385-86 (explaining that only one or two sellers have ever opted to purchase his services without the $100 upgrade to have their listing go to Realtor.com)). · Other customers wanted the full benefits of the Realcomp MLS and therefore paid for the additional expense of using an Exclusive Right to Sell listing. (CCPF 895, 1029, 1042, 1054-55). · As described by one discount broker, under the Website Policy, the service provided by Realcomp "is severely degraded" for Exclusive Agency listings by "really 1imit(ing)" the sellers' listings ''to not as much exposure as they would like to have." (CX 525 (Adams, Dep. at 78-79); CCPF 870). · Realcomp's Website Policy causes "sellers to either sell their home with less key exposure - thereby risking that the home wil not be sold or wil take longer to sell - or else purchase services that they do not want or need." (RX 154-A-72; CCPF 872-73). The Realcomp Policies therefore restrict competition in two ways: by excluding or restraining competition from discount brokers and collectively withholding from consumers services they desire. The natue of the Policies therefore supports the application of an abbreviated rule of reason analysis, particularly in light of Realcomp's market power. 3. Realcomp Has the Power to Restrain Competition Among Real Estate Brokers The evidence shows that Realcomp has market power, which gives it the ability to restrain competition among its members, competing real estate brokers. (CCPF 677-764). Market power is the ability to raise prices above those that would be charged in a competitive market, i.e., the ability to restrain competition. See, e.g., NCAA, 468 U.S. at 109; General -54­ Leaseways, Inc. v. National Truck Leasing Ass 'n, 744 F.2d 588,596 (7th Cir. 1984); (CCPF 677). Market power is assessed in defined relevant product and geographic markets, which frame the context for a restraint's competitive effects. A relevant product market is simply all products or services that are reasonable substitutes from a buyer's point of view. United States v. E.1. du Pont de Nemours & Co., 351 U.S. 377, 394-95 (1956). A relevant geographic market is the area in which consumers can practicably turn to obtain the relevant product or service. Tampa Elec. Co. v. Nashvile Coal Co., 365 U.S. 320, 327-28 (1961). A firm's high market share in the relevant market, plus the presence of barriers to entry, wil support a finding of market power. See, e.g., United States v. Microsoft Corp., 253 F.3d 34, 51-56 (D.C. Cir. 2001); Rebel Oil Co. v. At!. Richfield Co., 51 F.3d 1421, 1434 (9th Cir. 1995); FTC v. Staples, Inc., 970 F. Supp. 1066, 1081-82 (D.D.C. 1997). a. "Network Effects" and Multiple Listing Services Any assessment of the market power of an MLS must take into account "network effects." (CCPF 695-708, 721-33). Network effects exist where the value or quality of a service to one user increases as the number of other users of the same service increases. (CCPF 700). MLSs exhibit network effects because the more members there are in an MLS, the more valuable the MLS is to each user - more cooperating brokers provide more exposure for listing brokers, and more listing brokers means more inventory for cooperating brokers. (CCPF 700-01, 721). An MLS therefore exhibits network effects from both sides of the market, and these effects reinforce one another. (CCPF 701-03). As Respondent's economist admitted, more listing brokers attracts more cooperating brokers, and more cooperating brokers attacts more listing -55­ brokers. (Eisenstadt, Tr. 1530; CCPF 722). The presence of network effects in MLSs makes market share a good indicator of market power, and network effects act as a barrier to entry b. Market Definition There are two, related relevant product markets in this case. (CCPF 682-708). The first relevant product market is the market for residential real estate brokerage services; this is the market in which Realcomp's members compete. (CCPF 684-85). The evidence is clear that for the vast majority of home sellers, there are no reasonable substitutes to real estate brokerage services. (CCPF 686-93). The second reI event market is the market for the supply of multiple listings services to real estate brokers, which is the market in which Realcomp competes. (CCPF 684-85); see also Thompson, 934 F.2d at 1572 (in antitrust challenge to MLS rules, relevant product market was "the market for multilist services"). Multiple listing services are an input into the market for residential real estate brokerage services, and the presence of network effects results in the MLS being a necessary input. (CCPF 684, 695, 704). The evidence shows that the relevant geographic markets in this case are four counties in Michigan: Wayne, Oakland, Livingston, and Macomb counties. (CCPF 709-20). The evidence regarding market definition is unebutted. c. Realcomp's Market Shares Show that it Has Market Power Realcomp's market shares indicate that Realcomp has substantial market power. The presence of network effects in the market for multiple listing services means that market share is a good measure of market power and the importance of the MLS to a broker's abilty to compete. (CCPF 721-23). The value of an MLS with a high market share in a given geographic market -56­ wil be much greater to brokers (and home buyers and sellers) than the value of an MLS with a small market share. (CCPF 723). The greater the market share, the bigger the network effects, and the more important the MLS is to brokers. (CCPF 723). Network effects also act as a barrer to entry. (CCPF 724-31); see, e.g., Microsoft, 253 F.3d at 54-56 (describing barrier to entr in softare market due to network effects). They prevent competitor MLSs from easily expanding their share of listings. (CCPF 724-28). A shift in shares would require that both cooperating brokers and listing brokers simultaneously switch to the competing MLS. (CCPF 725-26). But a listing broker has little incentive to list a propert in an MLS with a small market share because there wil be few cooperating brokers searching that MLS for homes in the area. (CCPF 726). Similarly, a cooperating broker has little incentive to search an MLS with a small share oflistings. (CCPF 726). Successful entry by a rival MLS is therefore improbable because of high collective switching costS.24 (CCPF 726). To calculate Realcomp's market share, Dr. Wiliams used the listing data from Realcomp, MiRealSource, and all of Realcomp's data-sharing parners. (CCPF 738). Dr. Wiliams first "new listings" - the percentage of calculated Realcomp's share of listing that were placed on only one MLS. (CCPF 738). As Dr. Wiliams explained, however, market shares based on new listings may understate Realcomp's importance to brokers. New listings do not include listings 24 As explained by the industry expert, the fact that Realcomp has approximately 14,000 members - the largest in Southeastern Michigan - means that it wil be "more effective" at helping brokers serve their members because it provides "more potential cooperating brokers with more buyers to help sell your home (and i)fyou're a cooperating broker, you've got more inventory to look at." an MLS increases with the more listings it has (Murray, Tr. 182-183; RX 154-A-031 ("The value of because that increases the likelihood that brokers wil be able to match a wiling buyer with a wiling seller."); CCPF 517). In fact, an MLS with -57­ that brokers place on any MLS in addition to Realcomp; thus, even though the broker felt compelled to list on the Realcomp MLS, the share of new listings does not reflect this importance. (CCPF 744-46). Dr. Wiliams therefore calculated Realcomp's share of "unique" listings - the share of all listed homes that are listed on Realcomp (whether or not listed on another MLS). (CCPF 744-76). As Realcomp' s economist admitted on cross-examination, in markets with entr bariers, significant market power may be inferred based on market shares above 33% (and in some markets, shares above 25%). (Eisenstadt, Tr. 1529-30; CCPF 735). By any measure, Realcomp's market shares of . The data show that Realcomp has_ both new and unique listings: Wayne Oakland Livingston Macomb Al Four Counties (CCPF 739-41, 747-49). -58­ Maps showing these shares at the zip code level show how Realcomp' s market share spreads throughout each of the four counties: (CX 507, in camera; CX 514, in camera). Based on all of the analyses, Dr. Williams's unebutted conclusion is that Realcomp possesses substantial market power in the market for multiple listing services in Oakland, Wayne, and Livingston counties and to a lesser extent in Macomb county. (CCPF 757); see also Reifert v. South Cent. Wis. MLS Corp., 450 F .3d 312, 317 (7th Cir. 2006) (holding that MLS had market power because of high share of homes listed on service, the features and information on the MLS was not available through another service, and free listing services and newspapers were not a substitute for the MLS). -59­ d. Realcomp's Market Power in the Multiple Listing Services Market Gives Realcomp the Abilty to Restrain Competition in the Real Estate Brokerage Services Market The evidence shows that Realcomp's provision of means that it is a key input in the real estate brokerage services within Southeastern Michigan. (CCPF 757-64). This means that Realcom can restrict competition in the market for real estate brokerage services by hindering or excluding certain tyes of competitors, such as discount brokers. (CCPF 758-60); see also Realty Multi-List, 629 F.2d at 1370-71 (holding that unreasonably restrictive MLS rules imposed by an MLS with market power wil harm competition among brokers). The ability to restrct competition in the brokerage services market also means that the Realcomp can harm consumers. (CCPF 761-63); see also Realty Multi-List, 629 F.2d at 1371 (holding that buyers and sellers are also harmed by MLS rues that restrct competition among brokers); Thompson, 934 F.2d at 1580 (stating that uneasonably restrictive MLS rules wil result in "inefficient sales prices" and "could result in less competition for brokerage fees"). e. Other Evidence Confirms That Realcomp has Market Power and the Abilty to Restrict Competition Among Brokers Cours also have found market power based on evidence that an MLS has such economic importance that the broker's exclusion "results in the denial of an opportity to compete effectively on equal terms." Realty Multi-List, 629 F.2d at 1373; see also Thompson, 934 F.2d at 1580 (adopting Realty Multi-List standard); Northwest Wholesale Stationers, 472 U.S. at 296 (holding that group boycott would be subject to per se treatment if the cooperative "possesses market power or exclusive access to an element essential to effective competition"). -60­ Realcomp also has market power under this standard. The evidence uniformly shows that access to Realcomp' s services is critical to brokers' ability to compete effectively in the relevant market: · Placing a listing on the Realcomp MLS exposes the listing to the thousands of Realcomp members working with buyers, which is "great exposure for a house that's for sale." (Mincy, Tr. 318; CCPF 505). · Realcomp members practicing in Wayne, Livingston, Oakland and Macomb counties repeatedly testified that Realcomp is their local MLS, and as one Realcomp member stated to potential.clients - a broker "must" belong to their local MLS. (CCPF 511-12; CX 307-001; CX 421 (Whitehouse, Dep. at 46-48); CX 39 (Taylor, Dep. at 16-17) (testifying that it only makes sense to list properties on the local MLS)). · Both discount and full-service Brokers consistently testified about the importance ofRealcomp's feed oflistings to public and IDX websites. (CCPF 656-67). As Realcomp's witness, Mr. Sweeney explained at trial, "if a broker elected not to feed their listings to any Realtor.com, IDX feeds and so on and so forth, that probably would be business suicide. . . . it would definitely put them at a severe competitive disadvantage." (Sweeney, Tr. 1347; CCPF 667). · No other local MLS provides this exposure for brokers in Southeastern Michigan. (CCPF 513). * * * Under an abbreviated rule of reason analysis, the combination of the natue of Realcomp's Policies and Realcomp's market power establishes that Realcomp's Policies are presumptivelyanticompetitive. The natue of the harms caused by Realcomp's Policies are the same as those caused by other restraints subject to an abbreviated analysis. The evidence shows that Realcomp's Policies tend to exclude competition from discount brokers just like the exclusionary MLS rules addressed in Realty-Multi List and Thompson. Under those cases, the evidence of market power shifts the burden of proof to Realcomp to justify its Policies, see -61­ Thompson, 934 F..2d at 1581, without any need to show actual effects. See Realty Multi-List, 629 F.2d 1375; see also Northwest Wholesale Stationers, 472 U.S. at 296 (rule excluding competitor from services of cooperative may be per se ilegal if cooperative possesses market power); cf Indiana Fedn, 476 U.S. at 461-62 (holding that given nature of restraint and evidence of market power, restraint could be condemned even absent proof of actual effects); Brown Univ., 5 F.3d at 668 (proof of actual anticompetitive effects unecessary if market power shown). In fact, an abbreviated rule of reason analysis is even more appropriate in this case because the evidence shows that Realcomp' s Policies also restrict competition among Realcomp members through a competitor agreement to withhold desired services. The Supreme Court has held that this type of agreement is presumptively anticompetitive even without proof of market power. Indiana Fedn, 476 U.S. at 459-60; see also Cal. Dental Assoc., 526 U.S. at 770. Complaint Counsel has therefore shown that Realcomp's Policies are presumptively anticompetitive, and the Cour need not consider the evidence of actual competitive effects. In fact, the cours tyically do not require proof of actual effects even under a full rule of reason analysis because of the "diffculty of isolating the market effects of challenged conduct." Brown Univ., 5 F.3d at 668; Gordon v. Lewiston Hosp., 423 F.3d 184,210 (3d Cir. 2005) ("under the traditional rule of reason... (b )ecause proof that the concerted action actually caused anticompetitive effects is often impossible to sustain, proof of the defendant's market power wil suffce"); Flegel v. Christian Hosp., Northeast-Northwest, 4 F.3d 682,688 (8th Cir. 1993) (explaining that under the rule of reason "( e lither showing - market power or actual detrimental effects - shifts the burden to the defendant to demonstrate pro-competitive effects"). Nonetheless, the record is replete with evidence that Realcomp's Policies actully restrain -62­ competition. This evidence is more than suffcient to show that Realcomp's Policies are anticompetitive, even under a fuller market analysis. 4. Evidence of Actual Anticompetitive Effects Confirms that Realcomp's Policies Harm Competition and Consumers The record evidence shows that Realcomp's Policies are, in fact, effective. The Policies have substantially reduced limited service brokerage activity within the Realcomp MLS (and therefore within each of the four relevant counties). Furher, the evidence shows that Realcomp's Policies force consumers to purchase services they do not want. Finally, the evidence shows that the Policies likely help to sustain higher prices for real estate brokerage services. This evidence fuher substantiates that Realcomp has market power, see, e.g., Indiana Fedn, 476 U.S. at 460-61, confirms that the nature ofRealcomp's Policies is to restrict competition, and is suffcient to support the conclusion that Realcomp's Policies cause anticompetitive effects under any analysis. a. Realcomp's Policies Substantially Reduced Limited Service Brokerage Activity The empirical evidence shows that Realcomp's Policies reduced the share of Exclusive Agency listings within the Realcomp MLS. (CCPF 1069-1122). Dr. Wiliams's statistical analyses show that Realcomp's Policies are associated with a reduction in the share of Exclusive Agency listings of 5.5 to 5.8 percentage points. (CCPF 1102). Dr. Eisenstadt's two benchmark analyses imply reductions due to Realcomp's policies of 0.8 and 3.4 percentage points. (CCPF 1111-12). Even Dr. Wiliams's time-series analysis, which because of Realcomp's change in reporting is likely to understate the effect of Realcomp' s policies, shows a reduction of 0.9 percentage points. (CCPF 1081-83). -63­ The reduced share of Exclusive Agency listings shows actual anticompetitive effects. First, the reduction in the share of Exclusive Agency listings is competitively significant because brokers using these tyes of listings provide a different and important form of competition. (CCPF 1149-55, 1219-23). Limited service brokers "unbundle" listing services, allowing home sellers to purchase a subset of brokerage services, while "self-supplying" other services. (CCPF 1139). This unbundling allows brokers to offer substantial savings on the cost oflisting services. (CCPF 199-200, 1137-39). Moreover, by using Exclusive Agency contracts, limited service brokers unbundle commissions, allowing consumers to avoid paying the cooperating broker commission (i.e., the offer of compensation) ifno cooperating broker participates in the transaction. (CCPF 184,202-03). By competing in this way, the evidence shows that discount brokers put price pressure on traditional, full-service brokers. (CCPF 221-26). Second, the reduction in the share of Exclusive Agency listings, though small in absolute terms, is a substantial reduction in the extent oflimited service brokerage activity. (CCPF 1192­ 99). For instance, Dr. Wiliams's time-series analysis shows that the share of Exclusive Agency listings dropped from about 1.5% to 0.72%. (CCPF 1192). While the drop is only 0.79 percentage points, it represents a 52% drop in limited-service brokerage activity. Dr. Williams's benchmark and statistical analyses show that Realcomp's Policies reduced limited-service brokerage activity by 84% to 86%. (CCPF 1197-98). Similarly, Dr. Eisenstadt's two benchmark analyses using the Ann Arbor MLS show that Realcomp's Policies reduced limited-service brokerage activity by 55% and 82%. (CCPF 1196). All of the data therefore tell the same story: Realcomp's Policies are associated with a large reduction in limited service brokerage activity and the type of competition that activity brings. (CCPF 1199). -64­ b. Realcomp's Policies Have Forced Consumers to Purchase Unwanted Services and to Pay More for Brokerage Services Realcomp's Policies also force consumers to purchase unwanted services. By favoring Exclusive Right to Sell listings, Realcomp's Policies steer some consumers who would prefer Exclusive Agency listings to purchase Exclusive Right to Sell listings. (CCPF 1029, 1055, 1201). These consumers pay for the services of cooperating brokers, even when those brokers are not involved in the transaction. (CCPF 1228, 1230-31). Thus, by shifting consumers to Exclusive Right to Sell listings, Realcomp's Polices increase the cost of real estate services to consumers. (CCPF 1212-33). Realcomp's Policies further reduced consumer choice by requiring that brokers offer "full service" with Exclusive Right to Sell listings to obtain full exposure. (CCPF 1205, 1232). Consumers were therefore not only foreclosed from purchasing Exclusive Agency listings with full exposure, they also had to purchase all of Realcomp' s minimum services to have what Realcomp considered an Exclusive Right to Sell listing. (CCPF 182, 1233). c. Realcomp's Policies Have Likely Maintained Higher Brokerage Fees and Reduced Output of Brokerage Services Perhaps the most insidious effect of Realcomp's Policies, however, is that they help to maintain a price floor for real estate brokerage services. The economic evidence shows that because of the structue of an Exclusive Right to Sell agreement, the minimum price for such a compensation. (CCPF 1212-15). listing that wil cover a broker's expected costs is the offer of Under Exclusive Right to Sell agreements, the total commission is not contingent on whether a cooperating broker is involved. (CCPF 1214). The listing broker and the consumer negotiate the commission up front, and the listing broker builds into the commission the expected payment of -65­ an offer of compensation. (CCPF 1215). Under an Exclusive Right to Sell listing, therefore, the consumer must commit to paying the offer of compensation whether or not a cooperating broker is eventually involved. (CCPF 1212-15). The evidence shows that the offer must be near. to be competitive in Southeastern Michigan. (CCPF 1210-11). Thus, the minimum price for an Exclusive Right to Sell contract wil be at least this amount. (CCPF 1218). This is a "de facto" price floor. (CCPF 1218). The economic evidence shows that by restraining competition from limited service brokers, Realcomp's Policies likely protected this de facto price floor. (CCPF 1219-27). Taking steps to "prevent a price collapse" through a competitor combination has long been ilegaL. Toys "R" Us, Inc. v. FTC, 221 F.3d 928,937 (7th Cir. 2000). The evidence also shows that Realcomp's Policies have likely reduced the output of brokerage services. Limited service brokers are particularly appealing to for-sale-by-owner sellers. (CCPF 1237-38). According to Realcomp's own expert, Realcomp's Policies cause some sellers who would have purchased Exclusive Agency listings to decide to sell their homes own their own. (CCPF 1239). Realcomp's expert admitted on cross-examination that this meant that "as a consequence of the rule" there would be "less use of brokers." (Eisenstadt, Tr. 1486­ 87; CCPF 1239). Thus, to the extent that discount brokerage services are restrained, overall output of brokerage services is reduced because fewer FSBO sellers wil chose to convert to using brokerage services. -66­ d. The Evidence Establishes that Realcomp's Policies Have Had Substantial Actual Antiticompetitive Effects This evidence is more than suffcient to show actual anti competitive effects, even under a fuller market analysis. Given the choices and form of competition that limited service brokers provide, even a very small reduction in the absolute market share of Exclusive Agency listings shows anticompetitive effects. The Commission has previously held that even smaller reductions in the market share of emerging competitors show anticompetitive effects under a full rule of reason analysis. In Toys "R" Us, Inc., 126 F.T.C. 415 (1998), aff'd, 221 F.3d 928 (7th Cir. 2000), the Commission held that Toys "R" Us had restrained trade by organizing a horizontal agreement among competing toy manufactuers to not sell certain types of toys to wholesale clubs, "an innovative class of discount retailers" that competed with the respondent. Id. at 527-28. The evidence showed that the wholesale clubs accounted for 1.9% of total toy sales in 1992, but after Toys "R" Us orchestrated the restraint, the clubs' market share decreased to 1.4% in 1995, a 0.5 percentage point reduction in market share. Id. at 597. The respondent argued that the Complaint Counsel had failed to demonstrate anticompetitive effects because the clubs' market share was so low and there was no evidence of a market-wide price increase or reduction in output. Id. at 609,611. The Commission, however, held that Toys "R" Us was not "privileged to organize a boycott designed to disadvantage" the clubs merely because their market share was smalL. Id. at 609. Rather, the Commission concluded that the 0.5% reduction in the clubs' market share supported a finding of anticompetitive effects, especially in light of evidence that the clubs put price pressure on the -67­ respondent and that the restraint "reduced the range of choices available to consumers and eliminated forms of competition that consumers desired." Id. at 610. Furher, citing the Supreme Cour's decision in Indiana Federation of Dentists, the Commission explicitly rejected the argument that Complaint Counsel needed to prove market-wide price or output effects. Id. at 611. The very factors relied on by the Commission to find anticompetitive effects are present in this case. The evidence shows that the Realcomp Policies reduced the share of Exclusive Agency listings, and therefore limited service brokerage activity by more than the 0.5 percentage point decline in Toys "R" Us. (CCPF 1081, 1102, 1106). Limited service brokers represent a different and important form of competition, offering substantial cost savings to consumers and putting price pressure on traditional brokers. (CCPF 1149-55, 199-201,221-26). And Realcomp's Policies limit consumer choice by eliminating the product desired by consumers - an Exclusive Agency listing with full exposure through the Realcomp MLS. (CCPF 1124, 1200­ 06); see also Glen Holly Entertainment, 352 F.3d at 374,378 (holding reduction in consumer choice is anticompetitive har even without any price increase or output reduction). This evidence is therefore sufficient to show that Realcomp's Policies are anticompetitive, even under a fuller market analysis. 5. Realcomp's Policies Lack Any Procompetitive Justification At trial, Realcomp asserted that the purpose of the Website Policy is to ensure that sellers using Exclusive Agency listings pay commissions to Realcomp brokers (both listing brokers and -68­ cooperating brokers).5 (See Kage, Tr. 1050-52).26 According to Realcomp's witnesses, the problem being addressed is that if Exclusive Agency listings were sent by Realcomp to public websites, a buyer might find the listing and purchase the propert "without the services of a realtor." (Sweeney, Tr. 1333-34;27 Kage, Tr. 1051 (testifyng that sellers benefit from the advertising but "could sell the propert to somebody that includes no commission")). Realcomp's economist called this a tye of "free riding." (Eisenstadt, Tr. 1401-02 (testifying that a seller using an Exclusive Agency listing might "capture(J for themselves the three percent cooperating broker commission"). The Commission, however, has already rejected this asserted efficiency justification for a website policy. See, e.g., Analysis of Agreements Contaíning Consent Orders to Aid Public Comment, In the Matter of Information and Real Estate Services, LLC, File No. 06-10087 25 At trial, Realcomp's CEO claimed that the reason for the Search Function Policy was to default to only Exclusive Right to Sell listings because brokers had complained that when showing Exclusive Agency listings they had to deal with the seller directly instead of another broker. (Kage, Tr. 1038-39). This justification might satisfy some brokers, but it does not improve competition. Moreover, every listing on the Realcomp MLS is identified by listings type, allowing brokers to know in advance if they might have to deal directly with a seller. (CCPF 323). The policy is therefore not needed and overly broad. The Search Function Policy has no procompetitive justification. 26 While Realcomp elicited testimony from Karen Kage regarding the reason the Board adopted Governors could testify as to these reasons. . (CCPF 1266-80). Moreover, Ms. Kage's testimony is not supported by a single contemporaneous document. (CCPF 1281-85). Instead, the contemporaneous documents indicate that Realcomp wanted to exclude Exclusive Agency listings entirely. (CCPF 784-87, 818-21). Realcomp's supposed reasons for the Website Policy should therefore be viewed as post hoc rationalizations, which deserve no weight. See, e.g., United States v. Dentsply Intl, Inc., 399 F.3d 181, 197 (3d Cir. 2005) (alleged justification was pretextual and did not excuse exclusionary practices); Image Tech. Servs. v. Eastman Kodak Co., 125 F .3d 1195, 1219-22 (9th Cir. 1997) (allowing fact-finder to disregard justification for challenged conduct when "evidence suggests that the proffered business justification played no par in the decision to act"). the Website Policy, not a single member of the Board of 27 Mr. Sweeney has never been on the Realcomp Board of Governors, has not been on the Realcomp MLS user committee since at least 2000, and did not attend the Realcomp Board of Governors meetings where the Search Function or Website Policy were adopted. Mr. Sweeney testified to his personal opinions, and not on behalf of the Realcomp Board of Governors. (CCPF 64). -69­ (2006). As the Commission explained, website policies like Realcomp's "advance no legitimate procompetitive purpose": Exclusive Agency Listings do not enable home buyers or sellers to bypass the use of the brokerage services that the MLS was created to promote, because a listing broker is always involved in an Exclusive Agency Listing, and the MLS rules. . . already provide protections to ensure that a (cooperating) broker - a broker who finds a buyer for the propert - is compensated for the brokerage services he or she provides. Id. at 7. Applying the Commission's analysis, the evidence in this case shows that Realcomp's asserted justification is not procompetitive. Realcomp has already stipulated that (1) an "Exclusive Agency Listing involves the services of a listing broker" (JX 1-07 (i! 55); CCPF 183); (2) regardless of listing tye, "Realcomp does not require that brokers whose listings are transmitted by Realcomp to the Approved Web sites be compensated at all, whether by listings on the Realcomp commission or otheiwise" (JX 1-04 (i! 23); CCPF 373); and (3) all MLS "must make a unilateral offer of compensation" to cooperating brokers. (CCPF 350). Further, Realcomp's rules protect cooperating brokers' right to receive the offer of compensation if they procure a buyer for a propert. (CCPF 360, 362-63). Thus, on the Realcomp MLS, a listing broker is involved in every Exclusive Agency listing. (CCPF 183). That listing broker's commission is a matter of negotiation between the broker and the seller. (JX 1-04 (i! 26); CCPF 354). Ever Exclusive Agency listing includes an enforceable offer of compensation. (CCPF 230,358,360,363). Under established Commission analysis, therefore, the Realcomp Website Policy is not procompetitive. -70­ Moreover, the economic evidence shows that the Website Policy does not address any "free riding" concern. Free riding occurs when a customer uses the sèrvices of one seller and then makes a purchase elsewhere. (CCPF 1256). The classic example is when a customer goes to a full-service retailer, learns about the product, and then goes to a discount supplier to purchase the product. (CCPF 1256); see also Chicago Prof'l Sports Ltd. Partnership v. National Basketball Ass 'n, 961 F.2d 667,675 (7th Cir. 1992). The economic problem with free riding is that, in the long term, there wil not be sufficient incentives for the full-serice retailer to provide the servces. (CCPF 1256); Toys "R" Us, 221 F.3d at 937-38 ("What the manufacturer does not want is for the shopper to visit the attctive store with highly paid, intelligent sales help, learn all about the product, and then go home and order it from a discount warehouse or (today) on-line discounters. The shopper in that situation has taken a 'free ride' on the retailer's efforts; the retailer never gets paid for them, and eventually it stops offering the services."). There is no free riding here. (CCPF 1257-65). Consumers using Exclusive Agency listings do not free ride on listing brokers; the listing brokers are paid for their services. (CCPF 1258); Chicago Prof'l Sports, 961 F.2d at 675 ("What gives this the name free-riding is the lack of charge. . . . When payment is possible, free-riding is not a problem because the 'ride' is not free."). Nor do consumers using these listings "free ride" on cooperating brokers. (CCPF 1259). Cooperating brokers receive exactly what they pay for from the MLS: an opportity to earn a commission for bringing a buyer.28 (CCPF 1263). Lastly, consumers using Exclusive Agency 28 Moreover, the Website Policy allows Exclusive Right to Sell listings to be sent to public websites, and buyers may purchase homes under those listing agreements without a cooperating broker. (CCPF 1259). Disseminating these listings to public websites flies in the face of Realcomp's purported justification. (CCPF 1059). -71­ listings do not "free ride" on the MLS; the MLS is compensated for its servces by member fees, which all discount brokers pay. (CCPF 1264); see also Chicago Prof'l Sports, 961 F.2d at 675 (rejecting argument that member of competitor joint venture was free riding on the venture's advertising when venture could charge for this service). To the extent that the Realcomp Website Policy has any benefit, it benefits only traditional full-service brokers, not consumers. (CCPF 1265). This is not abstract reasoning. The National Association of Realtors - the organization whose purpose is to promote the interests of Realtors such as Realcomp's members - considered its hundreds and rejected Realcomp's arguments. NAR has made it mandatory for every one of of associated MLSs to include all listings, regardless of listing type, in any feed to public and IDX websites. (CCPF 839, 841-44). Realcomp tried to convince NAR not to do this, arguing the very same thing it does here - that allowing Exclusive Agency listings "is in direct conflict with the ver purpose of the MLS" because the seller could avoid paying a commssion. (CX 234-002; CCPF 846-47, 849, 852). But NAR, through its general counsel, squarely rejected Realcomp's arguments. (CX 234-002; CCPF 848, 850-51, 852-53). NAR explained that including Exclusive Agency listing on feeds to public websites and the IDX is not a problem because (1) "the seller had engaged the services of a real estate professional"; (2) these listings include "an offer of cooperation and compensation to MLS participants"; and (3) if a cooperating broker brings a buyer, "that broker is entitled to the compensation communcated to the MLS participants by the listing broker." (CX 234-002; CCPF 850-51, 852-53). Thus, concluded NAR, including Exclusive Agency listings on these feeds would not detract from the puroses of the MLS. (eX 234-003; CCPF 853). -72­ * * * The evidence shows that Realcomp violated Section 5 of the FTC Act. Realcomp's trade - Realcomp's Website members, competing brokers, combined to impose a restraint of Policy and its Search Function Policy. That restraint is unreasonable because it is likely to (and actually did) cause anticompetitive effects - reducing consumer choice, decreasing product quality, increasing prices, and lowering output. The only question that remains is the scope of the remedy that the Court should impose to restore and protect competition in the marketplace. iv. REMEDY Once an antitrust violation is established, the remedy must "unfetter (the) market from anti-competitive conduct and 'pry open to competition a market that has been closed by defendants' ilegal restraints. '" Ford Motor Co. v. United States, 405 U.S. 562, 577-78 (1972) (citations omitted)); see also United States v. United Shoe Mach. Corp., 391 U.S. 244, 250 (1968) (holding that the remedy must "terminate the ilegal monopoly, deny to the defendant the fruits of its statutory violation, and ensure that there remain no practices likely to result in monopolization in the futue"). In other words, it should "cure the il effects of the ilegal conduct, and assure the public freedom from its continuance." United States v. United States Gypsum Co., 340 U.S. 76, 88 (1950). A. Realcomp's Agreement to Change its Search Function Policy The parties' agreement to the entry of an order binding Realcomp to change its Search Function Policy is a start toward these goals. The stipulated order bars Realcomp from treating Exclusive Agency listings in a less advantageous manner than Exclusive Right to Selllistings -73­ with respect to the search function in the Realcomp MLS. Moreover, it gets rid of Realcomp's minimum services requirement for Exclusive Right to Sell listings. The agreement, however, explicitly does not address Realcomp's Website Policy. Realcomp therefore intends to continue to exclude Exclusive Agency listings from its feed of information to public websites and its IDX Further injunctive relief is stil needed. B. To Restore Competition, the Court Must Enjoin Realcomp from Enforcing Any Rules that Discriminate Against Brokers Using Exclusive Agency Listings The evidence shows that the continued existence of the Website Policy will restrain competition. First, the evidence shows that the Website Policy greatly impairs the ability to limited service brokers to market Exclusive Agency listings. (CCPF 868, 941-53, 1116-22). The policy effectively precludes these listings from three of the top four categories of real estate web sites used by buyers to search for listings, and makes it more costly for them to reach the other. (CCPF 875-85). Second, the data from other MLSs that imposed website policies but did not have anything like the Search Function Policy demonstrates that the Website Policy on its own reduces limited service brokerage activity. (CCPF 1089-1104, 1114-16). For instance, Realcomp's own economist demonstrated that the website policy in the Boulder MLS impacted the use of Exclusive Agency listings. (CCPF 1089, 1106). Third, limited service brokers consistently testified that changing the Search Function Policy would not resolve the competitive impact of Realcomp's Policies. (CCPF 970, 986, 992, 1000, 1004, 1021, 1025, 1028, 1042). The Court should therefore enter an order that includes the relief stipulated by the parties as well as relief that more broadly enjoins Rea1comp from treating Exclusive Agency listings, or any other lawful listing agreements, in a less advantageous maner than Exclusive Right to Sell -74­ listings. Because Realcomp violated Section 5, the Commission has broad authority to impose "fencing in" relief to ensure that Realcomp does not simply find some other means to accomplish the same restraint of trade. See FTC v. Ruberoid Co., 343 U.S. 470,473 (1952) ("the the precise form in which it is Commission is not limited to prohibiting the ilegal practice in found to have existed in the past. . . it must be allowed effectively to close all roads to the prohibited goal, so that its order may not be by-passed with impunity"). Even if there is any doubt that the Website Policy wil continue to restrain competition, "it is well settled that once the Governent has successfully borne the considerable burden of establishing a violation of law, all doubts as to the remedy are to be resolved in its favor." United States v. E. 1. duPont De Nemours & Co., 366 U.S. 316, 334 (1961). Here, Realcomp imposed two policies aimed at restraining competition from discount brokers using Exclusive Agency listings. The Cour should enter an order requiring Realcomp to cease and desist from all of its conduct. -75­ V. CONCLUSION Through their control ofRealcomp, the dominant local MLS, full-service brokers in Southeastern Michigan combined to restrain competition. The Website Policy and the Search Function Policy were effective. The Policies restrct competition from limited service brokers and withhold a product that consumers desire but full-service brokers dislike. This conduct violates Section 5 of the FTC Act. And the Cour should enjoin Realcomp to prevent fuher consumer har. Dated: August 3, 2007 Respectfully Submitted, ~(? Joel Chrstie Linda Holleran Sean P. Gates Peggy Bayer Femenella Chrstopher Renner Counsel Supporting the Complaint Bureau of Competition Federal Trade Commission Washington, D.C. 20580 Telephone: (202) 326-3711 Facsimile: (202) 326-3496 sgates~ftc.gov -76­ CERTIFICATE OF SERVICE This is to certify that on August 3, 2007, I caused a copy of the Public version of the attached Complaint Counsel's Post-Trial Brief to be served upon the following persons: by hand delivery to: The Honorable Stephen J. McGuire Chief Administrative Law Judge Federal Trade Commission 600 Pennsylvania Avenue, NW Washington, DC 20580 and by electronic transmission and overnight courier to: Scott Mandel, Esq. Steven H. Lasher, Esq. Foster, Swift, Collins & Smith P.c. 313 South Washington Square Lansing, M148933-2193 Counsel for Respondent Realcomp II Ltd.